Aftfc'

Mi

1

UNIVERSITY OF CALIFORNIA.

FROM THE LIBRARY OF

DR. JOSEPH LECONTE.

GIFT OF MRS. LECONTE.

No.

BY SPOTTISWOOPB AKD CO.
MSW-blliliET SQCiHB.

COSMOS:

SKETCH

OF A

PHYSICAL DESCRIPTION OP THE UNIVERSE,

BY

ALEXANDER VON HUMBOLDT.

I

VOL. IV.— PART I.

Natures vero rerum vis atquemajestas in omnibus momentisfide caret, si guis modo
varies ejus ac non totam complectatur animo.—Pmy. H. N. lib. vii. c. 1,

TRANSLATED T7NDEB THE SUPEBINTEKDENCE OF

MAJOE-GEKEEAL EDWAED SABINE, E.A., D.C.L.

V.P. AND TREAS. R.S.
CHEVALIEB OP THE ROYAL PBUSSIAN ORDER "pour le mtrite" IN SCIENCE.

LONDON:
LONGMAN, BROWN, GREEN, LONGMANS, & ROBERTS:

AND

JOHN MURRAY, ALBEMARLE STREET.

> 1858.
UNIVERSITY

EDITOR'S PREFACE.

As may be seen in page xxii. of the English translation
of the author's preface to vol. i., in p. 8 of vol. ii., and
in p. 13 of the present volume, it was the author's
intention that the fourth and concluding volume of the
" Kosmos " should contain, for the telluric portion of
the universe, a notice of the specialities of the several
branches of science of which the mutual connection
had been indicated in the " general view of nature "
presented in the first volume ; the specialities of the
uranological portion having been treated in volume iii.
It has proved impossible, however, to comprise within
the limits of an ordinary single volume the treatment
of both the " inorganic and organic domains " of ter-
.restrial nature, as contemplated in p. 13 of the present
volume, or even the whole of what is there indicated as
appertaining to the first of these divisions : the fourth

988S&

VI

volume will, therefore, consist of two parts, of which
the first was published in the Grerman original at the
commencement of the present year, and is now pre-
sented in the English translation; the possibility of
publishing the work in English within so short a time
of its appearance in German being the result of the
early possession of the larger portion of the proof
sheets, for which advantage the editor and translator
are indebted to the good offices of M. de Humboldt
himself, and to the obliging permission of the Grerman
publisher, Mr. Cotta.

The first 224 pages of the original text and notes
were printed early in 1854. The long attachment of
the illustrious author to the subject of terrestrial mag-
netism, which is contained therein, and which in a very
large measure owes to him, and to the impulse given by
him, the position which it now occupies, rendered him
even peculiarly desiious that its treatment should corre-
spond fully to the latest progress of our knowledge.
That progress has been such as to render the years
which have elapsed since 1854 equivalent to a much
longer interval in other departments of science. In
this view, therefore, besides making himself some brief
but important additions in pp. 449 to 452, M. de Hum-
boldt has expressed, on several occasions in the course
of our correspondence, a desire that I should make in

Vll

the English translation further rectifications and en-
largements, which he might afterwards embody in the
work itself, on the completion of its concluding portion.

In compliance with this honourable and gratifying
request, I have ventured to make some corrections in
pp. 104 to 107 of the original (114 to 117 of the
translation), and have added three notes — one on the
figure of the earth, and two on subjects in terrestrial
magnetism; these will be found in pp. 453 to 516.
That in so doing I have not gone beyond either the
letter or the spirit of M. de Humboldt's express desire
will, I think, best appear by the following extracts from
his correspondence, which I therefore permit myself to
make : —

" Tout ce qui appartient aux details des corrections
speciales, comme pp. 113-117, que je possede deja im~
primees en votre traduction, peut etre reserve a la
nouvelle edition de tout le ( Kosmos ' qui sera publiee
quelques mois apres que le 4me volume, seconde et
derniere partie, sera termine. Tous les changemens que
vous daignerez faire dans la traduction actuelle du 4me
volume, partie premiere (et je vous supplie d'en faire
beaucoup), seront employes consciencieusement." . . . " Je
ferai la plus vive attention a toutes les corrections que
vous daignerez aj outer en notes a la traduction, comme
aux changemens que vous introduirez immediatement

viii EDITOR'S PREFACE.

dans le texte. Je reserve tout cela pour des additions a
la fin du vol. iv, seconde et derniere section. C'est la que
trouveront leur places les changemens que je decouvre
dans vos pp. 113-117. Vous sentez, mon cher ami,
qu'il sera plus utile de reunir tout ce dont votre tra-
duction aura enrichi 1'ouvrage, et de le placer a la fin
de 1'ouvrage, en traduisant en allemand les passages
entiers quelques longs qu'ils puissent etre." . . . " Je veux
que rien ne se perde pour 1'Allemagne de ce qui vient
de vous, ayant la prevention de donner dans mon
( Kosmos ' vos travaux dans leur grand ensemble, comme
ils ne se trouvent encore nulle part."

Some passages which occur in the commencement of
the first volume of " Kosmos," relating to the physical
sciences generally, and to the manner in which, in
different stages of their progress, their results are more
or less susceptible of being presented under general
points of view, appear to me so appropriate to the
past and present state of the science of terrestrial
magnetism, that in attempting the following brief ex-
planation of 'the manner in which I have endeavoured to
fulfil the wish conveyed in the preceding extracts, I adopt
those expressions in great measure as conveying the views
which I entertain, and by which I have been guided.
Amongst the many branches into which terrestrial mag-
netism divides itself, it is only very recently that some

EDITOR'S PREFACE. ix

have begun to emerge from that state in which " facts,
though studied with assiduity and sagacity, still appear
for the most part unconnected, with little mutual rela-
tion, and it may be even in some instances in seeming
contradiction with each other." I venture to think that
these, the most advanced branches of terrestrial mag-
netism, have already in some measure reached the
stage at which " observations having multiplied, and
having been combined by reflection, more and more
points of contact and links of mutual relation are
discovered," and the intricacies arising from the " ex-
cessive combination of phenomena " are already yield-
ing to a " knowledge of the primary laws by which
they are regulated," which knowledge is, in its turn,
conducting to still " higher and more extensive general-
isations"— thus preparing the way for that yet more
advanced stage, when "a deeper insight into natural
forces may be attained." Yet while in regard to these
branches it is already becoming " more and more pos-
sible to develop general truths with conciseness without
superficiality," that task is still one of very great
difficulty, and admits of only imperfect and partial
realisation. Therefore, inasmuch as it was impossible
for me to comprehend within any admissible limits a
complete resume of all that has already been gained in
terrestrial magnetism, I have, on the present occasion,

written notes on two branches only of the general sub-
ject— viz. on the magnetic disturbances, and on the solar
diurnal variation of the declination ; in respect to these
branches " combination, by reasoning, of the aggregation
of observed facts," their " generalisation," and the pro-
gressive " discovery of laws," have advanced so far that
it has become possible, agreeably to the presiding idea
of M. de Humboldt's work on the Kosmos, to " arrange
the phenomena in such a connection and sequence as
may facilitate the insight into their causal connection."
To treat these two subjects " without superficiality " has
fully engrossed all the space that I could permit myself
to occupy, seeing the number of pages to which the
present volume has extended.

EDWARD SABINE.

13, Ashley Place, London :
April 14th. 1858.

CONTENTS.

EDITOR'S PREFACE

INTRODUCTION to the special results of observation in the

domain of terrestrial phenomena 3-12

FIRST SECTION 13-161

General remarks . . . . . . . .13-15

Magnitude, figure and density of the Earth . . . 16-33
Internal heat of the Earth and its distribution . . . 34-48
Magnetic activity of the Earth . . . . . 49-161

Historic portion 49-93

Intensity of the magnetic force .... 93-108

Inclination 108-125

Declination 125-153

Polar light or aurora 153-161

SECOND SECTION: —

Reaction of the interior of the Earth on its exterior . . 162-448

General remarks 162-166

Earthquakes (dynamic action) 166-184

Thermal springs 184-206

Springs of vapour and gas, salses, mud volcanoes,

and naphtha springs 206-221

Volcanoes . . . 221-448

Xll CONTENTS.

THE AUTHOR'S ANALYSIS or HIS DISCUSSION ON VOLCANOES.

Series of American volcanoes from 19^° north to 46° south lati-
tude ; Mexican volcanoes, pp 269 and 387 (Jorullo, pp. 289-
304, and Note 430) ; Cofre de Perote, Note 443 ; Cotopaxi, Note
544 ; Subterranean outbursts of vapours, pp. 320-323 ; Central
America, pp. 262-266, and Note 390 ^ New Granada and Quito,
pp. 269-273, and Note 393 (Antisana, pp. 310-317; Sangay,
p. 424 ; Tungurahua, p. 423 ; Cotopaxi, Note 454 ; Chimborazo,
Note 604); Peru and Bolivia, Note 398; Chili, Note 399;
(Antilles, Note 555).

Number of all the active volcanoes in the Cordilleras, p. 273 ;
Proportions of distances with and without volcanoes, p. 277;
Volcanoes in North-west America north of the parallel of the
Kio Gila, pp. 388-403 ; Review of all volcanoes not belonging
to the New Continent, pp. 326-386 ; Europe, pp. 320-328 ;
Islands of the Atlantic, pp. 328-332 ; Africa, pp. 332-334 ;
Asia, continent, pp. 334-348 ; Thian-schan, pp. 337-342, and
Notes 567-572 ; Kamtschatka, pp. 342-348 ; East Asiatic
Islands, pp. 349-361 (Island of Saghalin, Tarakai or Karafuto,
Note 485; Volcanoes of Japan, pp. 356-361); South Asiatic
Islands, pp. 361-367 (Java, pp. 280-289); Indian Ocean, pp.
367-371 ; Pacific Ocean, pp. 372-386.

Supposed number of volcanoes on the globe and their distribution
on continents and in islands, pp. 406-411 ; Distance of volcanic
activity from the sea, pp. 277, 412-413 ; Areas of depression,
pp. 412-415, and Note 567 ; Maars (mine-funnels), pp. 229-231 ;
Different modes by which, without the elevation or formation of
conical or dome-shaped frameworks, solid masses may arrive at
the surface of the Earth from its interior through fissures (basalts,
phonolites, and some pearl-stones ; beds of pumice seem also to
have proceeded not from summit-craters, but from such fissures).
Some streams from volcanic summits themselves consist not of a
connected fluid, but of disconnected scorise, and even of series of
expelled blocks and fragments ; there have been eruptions of
rocks which have not all been glowing, pp. 289, 310, 313-316,
323, 324.

CONTENTS. Xlll

Mineralogical composition of volcanic rocks ; generalisation of the
term "trachyte," p. 428 ; Classification of trachytes, in respect to
the association of their essential ingredients, into six groups or
divisions, according to the determinations of Gustav Rose ; and
geographical distribution of these groups, pp. 429-434 ; names
of andesite and andesine, pp. 428, 436, and Note 609. Besides the
characteristic ingredients of trachyte formations, there are also
non-essential ones, whose frequency or constant absence in
different but often neighbouring volcanoes deserve great atten-
tion, p. 437 ; mica, p. 438 ; glassy felspar, p. 439 ; hornblende
and augite, pp. 439, 440 ; kucite, pp. 440, 441 ; olivine, pp. 441,
442 ; obsidian (and debate respecting the formation of pumice),
pp. 443-448 ; Subterranean pumice-stone quarries, separate from
volcanoes, at Zumbalica, in the Cordilleras of Quito, at Huichapa,
in the Mexican highland, and at Tschegem, in the Caucasus,
pp. 319-323 ; Diversity of conditions under which the chemical
processes of volcanic action proceed in the formation of the
simple minerals, and their association in trachytes, pp. 437, 447,
448.

RECTIFICATIONS AND ADDITIONS ... . . . pp. 449-452

EDITOR'S NOTES „ 453-516

AUTHOR'S NOTES . „ i-clxxiii

INDEX , clxxv

COSMOS.

VOL. IV,

COSMOS:

A PHYSICAL DESCRIPTION OF THE UNIVERSE.

SPECIAL RESULTS OF OBSERVATION IN THE DOMAIN OP
TERUESTRIAL PHENOMENA.

INTRODUCTION.

IN a work of very extensive scope, in which facility of
comprehension and the production of a clear and distinct
general impression are especially aimed at, the structure of
the work, and the co-ordination and arrangement of the
several parts which compose the whole, are almost even
more important than the richness and abundance of the
materials. This is the more sensibly felt in the " Cosmos,"
or te Book of Nature/' where it is necessary to separate
carefully the generalisation of our theme, both in the ob-
jective view of external phenomena and in the reflection of
nature in the mirror of man's inner being, from the nar-
ration of the several results of observation. The first two
volumes of my work were devoted to the generalisation thus

B 2

•* INTRODUCTION.

spoken of, first in the contemplation of the universe as a
natural whole, and next in an endeavour to show how in
the course of centuries, at different periods in the history
of the human race, and in the most different regions of the
earth, mankind had progressively advanced towards a recog-
nition of the concurrent action of the forces of nature.
Although the arrangement in a significant order of the
descriptions of phsenomena is in itself adapted to lead us to
recognise their causal connection, yet it could not be hoped
that a general representation should appear fresh, animated,
and life-like, unless restricted within such limits as should
not permit the general effect to be lost, under the excessive
and too crowded accumulation of separate facts.

As in collections of geographical or geological maps,
representing graphically the configuration of land and sea,
or the characters and arrangement of the rocks at the earth's
surface, general maps are made to precede special ones ; so
it has appeared to me most fitting that in the Physical De-
scription of the Universe, its representation as a whole,
contemplated from more general and higher points of view,
should be followed by the separate presentation, in the two
last volumes, of those special results of observation on which
the present state of our knowledge is more particularly based.
These two last volumes, therefore (as I have already remarked,
Bd. iii. S. 4 — 9 ; Engl. p. 4—9), are to be regarded simply
as an extension and more careful elaboration of the general
representation of Nature, which constituted my first volume ;
and as the third was devoted exclusively to the uranological
or sidereal domain of the Cosmos, so the present and last
volume is designed to treat of the telluric sphere, or of the
phsenomena belonging to the globe which we inhabit. We

INTRODUCTION. 5

thus retain the highly ancient, simple,, and natural division
of creation into the Heavens and the Earth, as preserved to
us in the earliest monuments of all nations.

If already, in the course of the last volume, in passing
from the consideration of the heaven of the fixed stars — in
which countless suns shine either singly, or revolving round
each other, or are known to us only as constituting the faint
light of distant nebulae, — we felt the transition to our own
planetary system to be a descent from the great and universal
to the relatively small and special, the field of contemplation
becomes restricted within yet far narrower limits, in passing,
as we are now to do, from the totality of the varied solar
system to one only of the planets which circle round its
central luminary. The distance of the nearest of the fixed
stars, a Centauri, is still 263 times greater than the dia-
meter of the solar system taken to the aphelion of the comet
of 1680, and yet the latter distance is itself 853 times
greater than that of our Earth from the Sun. (Kosmos,
Bd. iii. S. 582; Engl. p. 261). These numbers (in which
the parallax of a Centauri is taken at 0"*9187) assign
approximately the distance of a comparatively near region
of sidereal space from the conjectured outermost limit of the
solar system, and the distance of this latter limit from the
Earth.

Uranology, which occupies itself with the contents of the
remote regions of space, still preserves its ancient prerogative
of affecting the imagination by the most powerful impres-
sions of the sublime, the inconceivable vastness of the rela-
tions of space and number which it presents, the recognised
order and law which govern the movements of the celestial
bodies, and lastly, from the tribute of our admiration called

6 INTRODUCTION.

forth by those conquests from the dominion of the unknown,
which have been achieved by observation and intellectual
research. The sense of this regularity and periodicity had
impressed itself so early on the minds of men, that we often
find it reflected in their forms of speech, indicating a refe-
rence to the preordained course of the heavenly bodies. To
this it may be added, that among the laws which men have
as yet been enabled to recognise in the material universe,
those which regulate the movements of bodies in celestial
space are perhaps the most admirable by their simplicity ;
being founded exclusively on the measure and distribution
of aggregated ponderable matter, and its powers of attraction.
The impression of the sublime arising from the sensibly vast
and immeasurable, passes, almost unconsciously to ourselves,
by virtue of the mysterious bond which links the sensuous
with the supersensuous, into a higher region of ideas. There
dwells in the image of the immeasurable, the boundless, the
infinite, a power which disposes the mind to a serious and
solemn tone, with which, as with the impression of what-
ever is intellectually great or morally exalted, emotion
mingles.

The effect which the occurrence of unwonted celestial
phenomena produces so generally and simultaneously on
entire masses of population, testifies the influence of this
association of feelings. The power exercised on more sen-
sitive minds by the simple aspect of the star-strewn canopy
of heaven, is further augmented by augmented knowledge,
and by the application of instruments, which, invented by
man, extend his visual powers, and with them the horizon of
his observation. The impression of the incomprehensible
immensity of the universe thus subjected to law and regu-

INTRODUCTION. 7

la ted order, calls forth the feeling of tranquillity and repose.
This feeling takes from the unsearchable depths of space, as
of time, that which to the excited imagination had otherwise
attached to them, of shrinking awe. In all regions of the
earth, man, in the impulse of his natural sensibility, has
praised the " calm repose of a star-light summer night."

If, then, immensity of space and magnitude of mass belong
especially to the Sidereal portion of the Physical Description of
the Universe, in which the eye is the only organ of contempla-
tion, on the other hand the Telluric portion has the more than
countervailing privilege, of offering a greater scientifically-
distinguishable variety in the multifarious elementary sub-
stances with which it is conversant. We are in contact with
terrestri;il nature through the medium of all our senses ; and
as astronomy, or the knowledge of moving luminous heavenly
bodies, has given occasion to the admirable augmentation
which has taken place in the brilliant domain of the higher
analysis and the wide range of optical science ; so, on the other
hand, it is the terrestrial sphere alone which, by the diversity
of substances, and the complicated action of the forces
manifested in those substances, has afforded the foundation
of chemistry, and of all those physical sciences which treat
of phenomena distinguishable from light- and heat-exciting
undulations. Each of these great divisions of the study of
Nature exercises, in virtue of the character of the problems
which it proposes for solution, a special influence on the
character of the intellectual labour to which it has given
rise, and to the enrichment of human knowledge which is
the fruit of that labour

All cosmical bodies, excepting our own planet, and the
aerolites which are attracted by it, are, so far as we can

8 INTRODUCTION.

recognise them, simply homogeneous gravitating matter,
without specific, or, what is called, elementary diversity of
substance. This simplicity, however, is by no means to be
regarded as belonging to the actual nature and constitution
of those distant orbs themselves ; it is founded solely on the
simplicity of the conditions of which the assumption suffices
for the explanation and prediction of their movements in
celestial space, and (as I have already more than once had
occasion to remark : Kosmos, Bd. i. S. 56 — 60, and 141 j
Bd. iii. S. 4, 18, 21—25, 594, and 626; Engl. Yol. i.
p. 50—54 and 126 ; Yol. iii. p. 4, 18, 21—25, 421, and
44 8 J on the exclusion of all direct and assured perception on
our parts of diversities of substance in the heavenly bodies.
We thus have presented to us for solution the great problem
of a system of celestial mechanics, subjecting all that is
variable in the uranological sphere solely to the doctrine of
the laws of motion.

Periodical changes in the appearance of the light reflected
from the surface of Mars do indeed indicate difference of
seasons and meteorological phsenomena, i. e. precipitations
occasioned by the cooling of the atmosphere of the planet
at its opposite poles in the opposite portions of its year
(Kosmos, Bd. iii. S. 513 ; Engl. p. 371). Guided by
analogy and connection of ideas, we may indeed infer from
hence the existence of ice or snow (and therefore of oxygen
and hydrogen, the constituents of water) in the planet
Mars ; as we may also infer the existence of different kinds
of rock in the erupted masses and flat annular plains of the
Moon ; but we cannot assure ourselves of the actual state of
tilings by direct observation. Newton only permitted him-
self to entertain conjectures as to the elementary constitution,

INTRODUCTION. 9

of the planets belonging to the same solar system, as we
learn from an important conversation held by him with
Conduit at Kensington (Kosmos, Bd. i. S. 137 and 407 ;
Engl. p. 1 22 and 389) . The uniform spectacle of apparently
homogeneous gravitating matter aggregated in celestial orbs
has struck the imagination of man in various ways, the most
remarkable instance being, perhaps, the myth which lends
to the soundless deserts of space the magic of musical tones
(Kosmos, Bd. iii. S. 437—439 ; Engl. p. 315—317).

In the all but infinite variety of chemically-distinct sub-
stances, and the manifestations of force which they exhibit, —
in the formative and productive activity of the whole of
organic nature, and of many inorganic substances, — and in
the changes which produce the never-ending appearance of
origination and destruction, — the order-seeking intellect,
ranging through the terrestrial d9main, looks, often unsatis-
fied, for simple laws of motion. In the Physics of Aristotle,
it is said, "the fundamental principles of all nature are
variation and motion ; whoso does not recognise these, does
not recognise nature'' (Phys. Auscult. iii. 1, p. 200, Bekker) ;
and, alluding to " diversity of substance," " difference of
essence," he terms motion, in respect to the category of
" qualitatives," " transformation," aXAoiwo-ie • a term very
different from simple " mixture,'' pi&s, and an interpene-
tration, winch does not exclude re-separation (De gener. et
corrupt, i. 1, p. 327).

The unequal ascent of fluids in capillary tubes; en-
dosmose, so active in all organic cells, which is probably a
consequence of capillarity; the condensation of gases in
porous bodies (oxygen gas in platinum under a pressure of
above 700 atmospheres, and carbonic acid gas in beech*

10 INTRODUCTION.

wood-charcoal, where more than a third of the quantity of
gas is condensed in a liquid form on the walls of the cells) ;
and the chemical action of " contact substances/' which by
their presence (catalytically) occasion or destroy combina-
tions without taking themselves any part therein, — all these
phsenomena teach that substances exercise upon each other,
at infinitely small distances, an attraction dependent on
their specific essences. Such attractions cannot be con-
ceived without motion excited by them, although escaping
our visual perception. In what relation this mutual mole-
cular attraction, viewed as a cause of perpetual motion
on the surface of the globe, and, it is highly probable, also
in its interior, may stand to the attraction of gravitation, by
virtue of which the planets, and the central bodies around
which they revolve, are in perpetual motion, is wholly
unknown to us. Even & partial solution of such a purely
physical problem would constitute the highest and most
glorious prize, which the combination of experiment with
intellectual reasoning could attain in such lines of research.
In the above allusions to molecular, and what is commonly
called Newtonian attraction, I have not been willing to
employ the latter term to designate exclusively the attraction
which prevails in the regions of space, extending to illimitable
distances, and acting inversely as the square of the distance.
Such an application of the word Newtonian appears to me
almost an injustice to the memory of that great man, who
already recognised both the manifestations of force, whilst
at the same time, as if anticipating future discoveries, he
attempted, in his appendix to his work on Optics, to at-
tribute capillarity, and the little that was then known of
chemical affinity, to universal gravitation (Laplace, Expos.

INTRODUCTION. 1 1

du Syst. du Monde, p. 384 ; Kosmos, Bd. iii. S. 22, and 32
Anm. 39 ; Eugl. p. 21, and vii., Note 39).

As in the visible world it is especially on the oceanic
horizon, that optical illusions often hold out to the expec-
tation of the discoverer the promise of new lands, which for
a time remains unrealised, so has it been on the bounds of
the ideal horizon, in the remotest regions of the intellectual
world, that to the earnest inquirer many promising hopes
have arisen and have again faded away. Great discoveries
in recent times are indeed suited to heighten expectation on
this subject; such are contact-electricity, — rotation-mag-
netism, which can be excited by substances either in a fluid
or a solid state, — the attempt to regard all affinity as a con-
sequence of the electric relations of atoms to a predominating
polar force, — the theory of isomorphous substances applied
to the formation of crystals, — many phsenomena of the electric
state of the living muscular fibre, — and the knowledge gained
of the influence of the height of the Sun (the temperature-
raising solar rays) on the greater or less magnetic suscepti-
bility of a constituent of our atmosphere, oxygen. When
we see new light dawning from a previously unknown group
of phenomena in the material world, we may the more
hopefully think ourselves on the verge of new discoveries,
if the relations of the new facts to those with which we
were previously acquainted, appear obscure or even con-
tradictory.

I have by preference adduced examples in which dynamic
actions of motive forces of attraction appear to open the
path by which we may hope to approach nearer to the solu-
tion of the problems of the original, invariable (and therefore
termed elementary) heterogeneity of substances (as oxygen,

12 - INTRODUCTION.

hydrogen, sulphur, potash, phosphorus, tin, &c.), and the
degree of their tendency to combine, or their chemical
affinity. Differences of form and composition are, however,
I here repeat, the elements of the whole of our knowledge
of matter ; they are the abstractions under or through which,
by means of measurement and analysis, we endeavour to
comprehend the whole of the material world. The detonation
of fulminates with a slight mechanical pressure, and the
still more violent explosion, accompanied by fire, of chloride
of nitrogen, contrast with the explosive combination of
chlorine and hydrogen gas on exposure to the direct inci-
dence of a solar ray, more especially the violet ray. Change
of substance, combination and decomposition, mark the
perpetual circuit of the elements in inorganic nature, as
well as in the animated cells of plants and animals. " The
quantity of the existing substances, however, remains the
same ; the elements only change their relative positions/'

Thus the sentence anciently enounced by Anaxagoras,
still holds good, that " that which exists in the universe
suffers neither augmentation nor decrease •" and that what
his contemporaries termed the perishing of things, was a
mere dissolution of previous combinations. It is indeed
true that the terrestrial sphere, inasmuch as it is the seat of
the only organic corporeal world accessible to our observa-
tion, appears a continual field of death and of corruption ;
but it is also true, that the great natural process of slow
combustion, which we term corruption, does not bring with
it any annihilation. The disengaged substances recombine
in other forms, which, by the forces residing in them,
become the means of calling forth fresh life from the bosom
of the earth.

ARRANGEMENT OF TELLURIC PHENOMENA. 13

B.

RESULTS OF OBSERVATION IN THE TELLURIC PORTION
OF THE PHYSICAL DESCRIPTION OF THE UNIVERSE.

IN endeavouring to bring an immeasurable mass of mate-
rials, consisting of the most multifarious objects, into the
desired subjection, — i. e. to arrange the phsenomena in such
a connection and sequence as shall facilitate the insight into
their causal connection, — it is necessary, in order to make the
representation at once clear and comprehensive, not to allow
particular details, especially in long-explored and mastered
fields of observation, to escape from the higher point of view
of cosmical unity. The telluric, as opposed to the uranologic
portion of the Physical Description of the Universe, naturally
divides itself into two parts : — the Inorganic and the Organic
domain. '£\\Q Jirst comprises the magnitude, figure, and
density of the terrestrial globe ; its internal heat, and electro-
magnetic activity; the mineralogical constitution of the
earth's crust ; the reaction of the interior of the planet on
its surface, acting dynamically as in earthquake movements,
and chemically as in the processes of the formation and
alteration of rocks ; the partial covering of the solid surface
by the liquid expanse of seas ; the outline and configuration
of the more elevated portions of the solid surface, forming
continents and islands; and the general, outermost, gaseous
envelope of the earth, the atmosphere. The second, or the

1'i GENERAL ARRANGEMENT OF

organic domain, will embrace, not the different animated or
vegetable forms themselves, as in a description of nature,
but rather their places in reference to the solid and liquid
parts of the earth's surface, or the geography of plants and
animals and the gradations of races and tribes distinguishable
in the specific unity of mankind.

This division into two domains also belongs in a certain
degree to antiquity. A line of demarcation was drawn be-
tween the elementary processes, change of form and transi-
tion of substances into each other, on the one hand, and the
life of plants and animals on the other. The distinction
between plants and animals (in the entire absence of any
means of augmenting the visual powers) (l) was made to
rest either solely on intuitive apprehension, or on the dogma
of self-nourishment (Aristot. de Anima, ii. 1, T. i. p. 412,
a 14 Bekker), and internal impulse or volition leading to
motion. That kind of intellectual conception which I have
called intuitive apprehension, or rather intuition, and still
more, the Stagirite's own peculiar acuteness in the fruitful
combination of ideas, led him to discern the apparent transi-
tion from the inanimate to the animate, from the elementary
substance to the plant, and even to the view that, in the pro-
gressively higher processes of formation, there might be
found intermediate gradations from plants to the lower
kinds of animals. (Aristot. de part. Animal, iv. 5, p. 681,
a 12 ; and Hist. Animal, viii. 1, p. 588, a 4 Bekker). The
history of organic nature (taking the word history in its
primary signification, — therefore in relation to earlier periods
of time, to the periods of the ancient Floras and Faunas)
is so intimately allied to geology, /. e. to the sequence of
the successive superimposed strata of the earth's surface,

TELLURIC PHENOMENA. 15

and to the chronometry of the elevations of lands and moun-
tains, that it has appeared to me preferable, for the sake of
the links connecting such great and widely-diffused phse-
nomena, not to make the otherwise very natural separation
of organic and inorganic a primary element of classification
in a work on the Cosmos. The object in this work is not
to treat subjects morphologically, but rather to view Nature,
and the active forces of Nat are, in their totality.

16 MAGNITUDE, FIGURE, AND

I,

Magnitude, Figure, and Density of the Earth— Internal Heat of the
Earth, and its Distribution — Magnetic Activity, manifesting itself
in variations of Inclination, Declination, and Force — Magnetic
Storms — Polar Light, or Aurora.

THERE is contained in all languages, though it may be
etymologically represented under different symbolical forms,
an expression equivalent to that of " Nature/' (and some-
times, as man is inclined to refer always primarily to the
seat of his own habitation, " Terrestrial Nature"), desig-
nating the result of the harmonious concurrent action of a
system of impelling forces, which are themselves only known
to us through their effects in the production of motion,
combination, and separation, and partially in the formation
of organic tissues (living organisms), which reproduce their
like. " Naturgefiihl," the feeling for, or sentiment of
Nature, is, in dispositions accessible to such impressions, the
vague, but exciting and elevating impression of this general
systematic action. Curiosity is first arrested by the rela-
tions in space and magnitude of our planet, a globular mass
of almost imperceptible minuteness in the immeasurable
universe. A system of concurrent activities uniting, or (by
polar action) separating, substances, supposes dependence of
each particle on the others, in the elementary processes of
inorganic formation, as well as in the elicitation and main-

DENSITY OF THE EAETH. 17

tenance of organic life. The size and figure of the ter-
restrial globe — its mass (i. e. the quantity of matter of
which it consists, and which, compared with its volume,
determines its density, and thereby, under certain conditions,
. its internal constitution, as well as the measure of its
attracting force) ; — are all connected with each other by an
interdependence, more distinctly recognisable and more
accessible to mathematical treatment than that which we
have as yet been able to perceive in the vital processes
above alluded to, in thermal currents, or in the telluric
conditions of electro -magnetism and chemical changes of
substances. Relations which, in complicated phsenomena,
we are not yet able to measure quantitatively, may yet exist,
and may be rendered probable on grounds of induction.

Although we cannot, in the present state of our know-
ledge, reduce to the same law the two kinds of attraction —
viz. that which acts at sensible distances (as the mutual
gravitation of the heavenly bodies), and that which acts at
distances immeasurably small (molecular or contact-attrac-
tion),— yet it is not on that account the less credible, that
capillary attraction, and the action of endosmose, so im-
portant in the ascent of sap and other juices and for the
whole of vegetable and animal physiology, may be affected
by the amount and relations of gravity, as may also electro-
magnetic processes and chemical action. We may assume,
to take extreme circumstances, that if our planet had only
the mass of the Moon, so that the force of gravity at its
surface should have only about one-sixth of its present
intensity, the meteorological processes of our climates, the
hypsometrical relations of our mountain- chains, and the
physiognomy (facies) of our vegetation, would all be very

VOL. IV. *

18 MAGNITUDE, FIGURE, AND

different from what they are. The absolute size of our globe,
with which we are about to occupy ourselves, becomes im-
portant in the general economy of nature, on account of the
proportion subsisting between it and the mass and velocity of
rotation ; for, speaking generally throughout the universe, if
the dimensions of planets, the quantity of substance, or mass,
of each, and their respective velocities and distances, were all
to be increased or decreased in one and the same proportion,
then in this ideal Macro- or Micro-cosmos, all phenomena
dependent on relations of gravitation would remain un-
changed. (2,

a. Magnitude, Figure (Ellipticity) , and Density of the
Earth.

(Extension of the " Picture of Nature" in Kosmos, Bd. i.
S. 171—178 and 420—425. In the English, Vol. i.
p. 154—161 and 400—405.)

The Earth has been measured and weighed to obtain its
exact form, density, and mass. The precision which has
been constantly aimed at in these terrestrial determinations
has at the same time benefitted astronomy by the improve-
ments in measuring instruments and in methods of analysis
which the pursuit has called forth, no less than by the solu-
tion of the problem itself. Indeed, a considerable part of
the operation of the measurement of degrees is itself astro-
nomical ; altitudes of stars determine the curvature of the arc
of which the length is found by geodesical operations.
Methods have been discovered, in the higher branches of
mathematics, for obtaining from given numerical elements
the solution of the difficult problems of the form of the
Earth,— the figure of equilibrium of a fluid homogeneous

DENSITY OF THE EARTH. 19

mass, or that of a solid sli ell-like non-homogeneous mass,
rotating round a fixed axis. Erom Newton and Huygens,
the most celebrated geometricians of the eighteenth century
were occupied with this solution. Here, as always, it is
well to remember, that whatever is achieved by intellectual
effort and mathematical research, derives its value, not alone
from that which is actually discovered and added to the
domain of human knowledge ; but also, and more especially,
from the higher perfection and power to which the analytical
instrument has been wrought in the course of the investi-
gation.

" The geometrical, as distinguished from the actual phy-
sical, figure of the Earth, (3) is determined by what would be
the surface of water in a network of canals everywhere
covering the earth in connection with the ocean. This
ideal geometric surface (the extension and completion
of the oceanic surface) is everywhere perpendicular to the
direction of the forces compounded of all the attractions
proceeding from the several particles of which the earth is
composed, combined with the centrifugal force corresponding
to its velocity of rotation. (4) This figure can only be
regarded as approximating, on the whole, very nearly to
that of an elliptic spheroid of revolution ; for irregularities
in the distribution of the mass in the interior of the earth
produce, not only local variations of density, but also
irregularity in the geometric surface, which is the product
of the concurrent action of unequally distributed elements.
The physical surface of the earth is that of the actually
existing land and water." Geological reasons render it not
improbable that accidental alterations which may take place
in the molten materials in tip intrrior of the earth, easily

20 MAGNITUDE, FIGURE, AND

mobile notwithstanding the great pressure to which they
are subject, may cause internal displacements of mass, which
may modify, after very long intervals of time, the geometric
surface itself in the curvature of the meridians and parallels
within small distances; while the physical surface is ex-
posed, in its oceanic portion, to a constantly recurring
periodical displacement of mass by the ebb and flow of the
tides. The smallness of the effect on gravitation of the
first-named supposed class of phsenomena, may cause a very
slow and gradual change taking place in continental regions
to escape discovery by actual observation : according to
EesseFs calculation, in order to increase the height of the
pole at any particular place only 1", there must be supposed
to be displaced in the interior of the earth a mass of such
weight as that, its density being taken as equal to the mean
density of the earth, its bulk shall be equal to 114 cubic
geographical (German) miles. (German geographical miles
are 15 to a degree). (5) Large as this volume may appear
for the supposed displaced mass when we compare it with
the volume of Mont Blanc, of Chimborazo, of Kinchinjunga,
it will seem less so when we recollect that the terrestrial
spheroid contains above 2650 millions of such cubic
miles.

The problem of the figure of the Earth — the connection of
which with the geological question respecting an earlier fluid
state of the rotating planetary body had already been recog-
nised at the great epoch (6) of Newton, Huygens, and Hooke,
— has been attempted to be solved, with unequal success, in
three different ways : by astro-geodesical measurement of
degrees, by pendulum experiments, and by the inequalities
of the Moon in latitude and longitude. The first method

DENSITY OF THE EARTH. 21

divides itself in its application into two : viz. the measure-
ment of degrees of latitude on an arc of the meridian, and
the measurement of degrees of longitude on different
parallels. Although seven years have now elapsed since I
included in my " General Representation of Nature" the re-
sults of Bessel's important memoir on the dimensions of the
Earth, yet his investigation cannot even now be replaced by
a more comprehensive one, based on later measurements of
degrees. There are, indeed, to be soon expected, one
important addition, and one more perfect revision, — the
publication of the nearly completed Russian Arc, extending
almost from the North Cape to the Black Sea; and the
careful comparison of the standard employed in the Indian
Arc, whereby the results of the latter will be more assured.
By the determinations published by Bessel in 1841, the
mean dimensions of our planet, according to the most exact
investigation (7) of ten measured arcs, are as follows : the
semi-major axis of the spheroid of rotation, which approxi-
mates most nearly to the irregular figure of the Earth, is
3272077,14 toises; the semi-minor axis, 3261139^33 ; the
length of a quadrant of the Earth, 5131179t,81 ; the length
of a mean degree of latitude, 57013t,109; the length of a
degree of longitude, on the equator, 57108^520, and in
latitude 45°, 40449t,371; the ellipticity, or flattening at
the poles, 29-~r2 ; and the length of a German geographical
mile of 15 to an equatorial degree, 3807fc,23. The following
Table shows the increase of the length of a degree of the
meridian from the equator to the poles, as found by ob-
servation,— modified, therefore, by local disturbances of
attraction : —

22

MAGNITUDE, FIGURE, AND

£ •§

A

a
>

H

JSL, ii.vi i

«

H

§ S §8* ^ rrt3 ~ S'-- ^ . S

A

0

&DV3 r-'fflo 3 ti2^s"3
«^! ~-2 ^ •J3-^ "^Sa^4^^

C3 5* ^>^"'c/1 ** 5 **O *^ *^ ^% KC"t2 ^

lllllj IglljJ^III

'•=.•2 S •

, -N ^-^-' v-^»/

iflf'j

&sjsf

COOOOO-lr— I-^OCO 1O .COCOO «O «O»O

-g cs -^ sr1

SiSfclJ

o|^:-|
^sso *

1O rH C^ l.O CO CO i-O rH O3 CS CO OS 4jl »O Ol

OSOCO"^JSCMl~-.t^ rH COf-lO CC COCO

rHOir-^^HO^-HOO O COt^C^ CO OCS

"br «> '£5 o> <u

gS^Ji

8

COTJ.CS CO rHK5 ^- t^^p UO 1CC-

s

5 osocooscot-coco cxi ireocb co t-4fi

!§§

rH CO -^ 01 0 0 rH 03 rH T? Tp ^ ^ CO

51

^ wihOMW°^S 01 S&CO ^ 2cO

— s

fs

O i— lOCOrHr— (O3COO3 O3 i— llOrH CO rHCO

J

11

10 t-<0 ^ 10 1000 TP

Isi

5 ot^»oococb>«cs oj OrHo o oo

gas

If!

OOiCOCCCO03»003 rH O3COO3 rH COCO
C*rH »0 COCO IQrH COCOrH-^(

P*S S

« « 55

&rS a

«OeOCO-<*CCOJOJOJ •* CSCO01 rH CO 10
CO«O»OlOlOlOlfjiO rji CO i— 1 r- i COCO

o°

. • v^-/ ., . ,^^

"^ cL .

r3 -^

S

rrt

2

S3 §

O

M " O

o

| .S-sl = 11 8 2 o ^

Illlll III ll

The determination of the figure of the Earth by the mea-
surement of degrees of longitude in different parallels of

DENSITY OF THE EARTH. 23

latitude, requires great exactness in the observed differences
of longitude. Cassini de Thury and Lacaille, as early as
1740, availed themselves of powder signals in the measure-
ment of an arc perpendicular to the meridian in the parallel
of Paris. At a more modern period, the length of parallel
arcs, and the differences of longitude, were measured with
far better instrumental means, and with greater certainty, in
the course of the great Trigonometrical Survey of England ;
the determinations were between Beechy Head and Dunnose,
and between Dover and Ealmouth,(8) the differences of
longitude being indeed only 1° 26' and 6° 22' respectively.
The most brilliant operation of this kind was undoubtedly
the measurement of the arc between the meridians of
Marennes on the west coast of France, and Fiume. It
crosses the most western chain of the Alps, and the
Lombard plains of Milan and Padua : the measurement was
executed by Brousseaud and Largeteau, Plana and Carlini,
and extends over a direct distance of 15° 32' 21", almost
entirely under the middle parallel of 45°. The many pen-
dulum experiments which were made in the neighbourhood
of the mountains confirmed in a remarkable manner the
previously recognised influence of local attraction, shown
by the comparison of the astronomical latitudes with the
results of the geodesical measurements.^)

Next to these two classes of direct measurements of de-
grees, (a) of meridian al, and (b] of parallel arcs, mention
should be made of a purely astronomical mode of determining
the Earth's figure. It is founded on the influence exercised
by the Earth on the motion of the Moon (i. e. on the
inequalities in her latitude and longitude). Laplace, who
first recognised the cause of these inequalities, also indicated

24 MAGNITUDE, FIGURE, AND

the mode of applying them to this question, and sagaciously
remarked, that this method lias a great advantage which
detached measurements of degrees and pendulum experi-
ments cannot have, in giving in a single result the
mean figure of the Earth (or the form belonging to the
entire planet). I recall with pleasure the happy expressions
by which this method was characterised by its inventor, (10)
"that an astronomer, without quitting his observatory,
might learn from the motions of a single heavenly body the
precise form of the Earth which he inhabits." According
to a final revision of the inequalities in latitude and
longitude of our satellite, and the employment of several
thousand observations by Burg, Bouvard, and Burckhardt,(n)
Laplace found for the figure of the Earth an ellipticity of
•g-J-g-th, which approximates very nearly to that given by the
measurements of degrees, viz. -^-9.

The oscillations of a pendulum offer a third method of
determining the figure of the Earth (i. e. the ratio of the
major to the minor axis, under the assumption of the form
being that of an elliptic spheroid), by finding the law
according to which the force <jf gravity increases in pro-
ceeding from the equator to the poles. The oscillations of
a penduhm had been first applied to the determination of
time by the Arabian astronomers, and in particular by
Ebn-Junis, at the end of the tenth century, in the brilliant
period of the Abasside Caliphs ; ( 12) and after being neglected
for six centuries, were again employed by Galileo and by
Kiccioli at Bologna. (13) . By combination with wheel-work
for regulating the march of time-pieces (employed first in
the imperfect attempts of Sanctorius at Padua in 1612, and
subsequently in the finished work of Huygens in 1656), the

DENSITY OF THE EARTH. 25

pendulum became, in Eicher's comparison of the march of
the same astronomical clock at Paris and at Cayenne (in
1(572), the first experimental proof of the variation in the
force of gravity in different latitudes. Picard had, indeed,
been occupied in the preparations for this important expe-
dition, but he does not on that account attribute to himself
the merit of the first proposition. Richer left Paris in
October 1671, and Picard, in the description of his mea-
surement of a degree, published in the same year (1671),
remarked simply (14), that " at a meeting of the Academy,
one of the members expressed an opinion, that, on account
of the Earth's rotation, weights might be found to be less
heavy, or have less gravity, at the equator than at the pole."
He adds doubtingly, " that, indeed, according to some
observations made in London, Lyons, and Bologna, it would
seem as it the seconds pendulum required to be shortened
in approaching the equator : but that, on the other hand,
he is not sufficiently convinced of the accuracy of those
measurements, because, at the Hague, the length of the
seconds pendulum was found to be quite the same as at
Paris, notwithstanding the difference of latitude." "We
unfortunately cannot learn when Newton first obtained the
knowledge, which was to him so important, of Eicher's
pendulum results, obtained in 1672 but first published in
a printed form in 1679 ; or of Cassiai's discovery, in 1666,
of the ellipticity of Jupiter; or, at least, we cannot learn it
with the same exactness and certainty which we possess in
regard to his very late knowledge of Picard's arc. At a
period when, with so happy an emulation, theoretical views
stimulated to the undertaking of observations, and the
results of observation reacted in their turn on theory, the

26 MAGNITUDE, FIGURE, AND

exact knowledge of particular dates has a great interest in
the history of the establishment of physical astronomy on
mathematical foundations.

If the direct measurements of arcs of meridians and pa-
rallels (particularly the Erench arc (15) of the meridian
between lat. 44° 42' and 47° 30', and the parallel arc
between points situated east and west of the Graian, Cot-
tian, and Maritime Alps), (16) show great deviations
from the mean ellipsoidal figure of the Earth, — the irregu-
larities in the measure of the ellipticity given by pendulum
stations differently distributed or grouped as respects
geographical relations, are much more striking. The de-
termination of the figure of the Earth by the increase or
decrease of gravity (intensity of attraction at the respec-
tive places) supposes that the intensity of the force at the
surface of the terrestrial rotating spheroid has remained the
same as at the period of solidification from a fluid state,
and that no subsequent changes of density have taken place
init.(17) Notwithstanding the great improvement of instru-
ments and methods by Borda, Kater, and Bessel, there are at
present in both hemispheres, from South Shetland to Spitz-
bergen— therefore from 62° 56' S. to 79° 50' N. latitude-
only between 65 and 70 points very irregularly distributed
over the earth's surface, (ls) at which the length of the
simple pendulum has been determined with the same exact-
ness as the position of the place in latitude, longitude, and
height above the sea.

The pendulum experiments made on the part of an arc of
the meridian measured by the Erench astronomers, as well as
the observations made by Captain Kater in connection with
the Trigonometrical Survey of Great Britain, showed that the

DENSITY OF THE EARTH. 27

results could by no means be represented in every case by a
variation in the force of gravity in the ratio of the square of
the sine of the latitude. The English government decided,
therefore (at the recommendation of the Yice-President of
the Koyal Society, Davies Gilbert), on sending out a scien-
tific expedition for the extension of the inquiry at stations
far more widely removed from each other, which was entrusted
to my friend Edward Sabine, who had previously accompanied
Captain Parry, on his first Arctic expedition of discovery, as
astronomer. He visited in the years 1822 and 1823, for
the purpose of pendulum experiments, the West Coast of
Africa from Sierra Leone to the Island of St. Thomas near
the equator, Ascension, the coast of South America from
Bahia to the mouth of the Orinoco, the West Indies, and
New York, and subsequently the high Arctic north as far
as Spitzbergeri, and a previously unvisited part of East
Greenland, concealed behind formidable icy barriers, in
lat. 74° 32'. This brilliant and successfully executed under-
taking had the advantage of being directed to a single sub-
ject of research as its principal object, and of comprehending
points extending over 93 degrees of latitude.

The field of observation along the French arc was indeed
less near to the equinoctial and Arctic regions, but it had the
great advantage of a linear distribution of the places of obser-
vation and of direct comparison with the curvature found by
the geodesical and astronomical operations. Biot, in 1824,
continued the series of pendulum experiments from Eormen-
tera in 38° 39' 56" lat., where he had previously observed
with Arago and Chaix, to Unst, the northernmost of the
Shetland Islands, in 60° 45' 25"; and in conjunction with

28 MAGNITUDE, FIGURE, AND

Mathieu made similar determinations on the parallel of
Bourdeaux, Figeac, and Padua, to Fiume.(19) These pen-
dulum results, taken together with those of Sabine, give for
the whole northern quadrant the ellipticity of -g-^-g- ; but if
divided into two parts, the results are far less accordant, (20)
*. e. from the equator to lat. 45° -g-fg-, and from lat. 45° to
the pole -s-i-B- The influence of the presence of the denser
kinds of rock, such as basalt, greenstone, diorite, and mela-
phyre, as compared with the specifically lighter sedimen-
tary and tertiary formations, as well as the effect of volcanic
islands, (21) in increasing the intensity of gravity, has been
recognised in most cases in both hemispheres ; but there
still remain anomalies which are not wliolly explained by
the geological character of the rocks accessible to our
observation.

For the southern hemisphere we possess a small series of
excellent observations, thinly scattered, however, over a wide
extent, by Freycinet, Duperrey, Fallows, Liitke, Brisbane
and Eumker, [and Foster] . They confirm what had already
appeared so strikingly from the observations in the northern
hemisphere, that the intensity of gravitation is not the
same at all places having the same latitude ; so much so,
that the increase of gravity from the equator to the poles
would appear to be subjected to unequal laws under dif-
ferent meridians. Lacaille's pendulum experiments at the
Cape of Good Hope, and those made in the Spanish voyage
of circumnavigation of Malaspina, might have caused it to
be believed that the southern hemisphere was in general
considerably more flattened than the northern one; but, as
I have already stated, (22) more exact results at the Falkland

DENSITY OF THE EARTH. 29

Islands, and in New Holland, compared with New York,
Dunkirk, and Barcelona,, have proved the contrary.

From what has been said, it appears to follow, that the
pendulum, — (no unimportant instrument of geological in-
vestigation, being a kind of sounding-line by which the
deep unseen terrestrial strata may be reached), — affords us
less -well-assured knowledge respecting the figure of our
planet than that which may be gained from measurements
of degrees and the lunar movements. Strata concentric,
elliptic, each taken separately homogeneous, but increasing
in density according to certain functions of the distance in
proceeding from the surface towards the centre of the earth
(which are the supposed theoretical conditions), may, by
their actual diversity of constitution, position, and order of
density in different parts of the earth, occasion at the sur-
face local deviations in the force of gravity. If the circum-
stances which produce these deviations are much more
recent than the hardening of the outer terrestrial crust, we
may conceive the figure of the earth's surface as not having
been locally modified by the internal movement of the
molten masses. At all events, the diversities of the results
of measurements by means of the pendulum are far
too great to be now ascribed to errors of observation.
When pendulum .stations are grouped or combined in
various ways, so as to give accordance or systematic regu-
larity in the general results derivable from such com-
binations, they are still always found to give a greater
amount of ellipticity than measurements of degrees have
hitherto done.

If we take the result of these latter as now most generally
received, according to Bessel's last determination, viz. :

30 MAGNITUDE, FIGURE, AND

; the increase of the radius of the earth at the
equator, (23) in comparison with that at the poles, is the
difference between 3272077 and 3261139 toises ; being
10938 toises, or 65628 Paris, or 69943 English, feet, or
not quite 11 \ geographical miles. As it has been customary
to compare this equatorial convexity of the earth's surface
with well-measured mountain-elevations, I select as objects
of comparison the highest known summit of the Himalaya,
Kinchinjunga, 4406 toises, 26436 Paris or 28178 English
feet, as measured by Colonel Waugh, and that part of the
Plateau of Thibet which is nearest to the Sacred Lakes, Rakas-
Tal and Manassarovar, and which, according to Lieut. Henry
Strachey, attains the mean elevation of 2400 toises, or about
15347 English feet. It would follow from hence that the
enlargement of our planet in its equatorial zone is not
three times as great as the elevation of the summit of the
highest terrestrial mountain above the level of the sea, nor
five times as great as the general elevation of the Eastern
Plateau of Thibet.

This is the proper place for remarking that differences
between the results obtained for the amount of the earth's
ellipticity by measurements of degrees taken alone, and by
the combination of measurements of degrees and pendulum
experiments, are actually far smaller than we might be
inclined to suppose at the first sight of the fractions in
which those results are expressed. (24) The difference
between -g-J-o- and -^l-o- as the extreme results for the in-
equality of the equatorial and polar axes, is little more than
6600 Paris (7034 English) feet: not twice the height of
such small mountains as Vesuvius and the Brocken.

DENSITY OP THE EARTH. 81

As soon as the measurements of degrees in very different
latitudes had manifested that the interior of the earth
cannot be assumed to be of uniform density, since their
results showed an ellipticity much inferior to that assumed
by Newton (T-}-o), and greatly exceeding -yj-g-, the amount
corresponding to the hypothesis of Huygens of the whole
attraction being concentrated at the centre of the earth, —
the connection of the amount of ellipticity with the " law of
density" in the interior of the terrestrial spheroid could not
but become an important subject for analytical investigation.
Theoretical speculations on gravity led early to the conside-
ration of the attraction which might be exercised by moun-
tain masses rising like banks from the dry bottom of the
atmospheric ocean. Newton had already examined, in his
" Treatise of the System of the World in a popular way,
1728," how much a mountain 2500 Paris feet high, and
having a diameter of 5000 such feet, would deflect the
plumb line from its vertical direction. It was probably this
examination which gave occasion to the not very satisfactory
experiments of Bouguer at Chiinborazo,(25) and of Maske-
lyne and Hutton at Schehallien in Perthshire, near Blair
Athol ; to the comparison of the lengths of the seconds
pendulum on the sea-shore and at 6000 Erench feet above
the sea (at the Hospice on Mount Cenis by Carlini, and
near Bourdeaux by Biot and Mathieu) ; and lastly, to the
delicate and alone decisive experiments of Reich and Baily,
made with the ingeniously -devised " Torsion Balance,"
invented by John Mitchell, (26) and transmitted through
Wollastou to Cavendish.

These three different modes of determining the density of

32 MAGNITUDE, FIGURE, AND

our planet have already been so far treated of in detail,
(Kosmos, Bd. i; S. 176—178 and 424, Anm. 6 ; Engl. p. 159
— 160 and 404), that it will be sufficient to subjoin here the
result of the fresh experiments by Reich in 1847 and 1850,
published in a later memoir by that indefatigable investi-
gator^27) According to the present state of our knowledge,
the results of the three methods are the following; the
density of water being taken as unity.

The Schehallien experiments ; mean of Playfair's

maximum 4*867, and his minimum 4*559 .... 4*713
Pendulum observations at Mont Cenis, by Car-

lini, with Giulio's correction 4*950

Torsion Balance; Cavendish, according to

Baily's calculation 5*448

Reich, 1838 5*440

Baily, 1842 5*660

Reich, 1847—1850 5*577

The mean of the two last results gives the density of the
earth 5 '62; much exceeding, therefore, that of the densest
and finest-grained basalts (according to Leonhard's nume-
rous experiments, from 2*95 to 3*67) ; exceeding that of
magnetic iron ore (4'9 — 5'2) ; and but little inferior
to the native arsenic of Marienberg or Joachimsthal.
I have already remarked (Kosmos, Bd. i. S. 177; Eng-
lish edition, p. 160), that, viewing the great proportion of
the visible strata of our continents which are secondary,
tertiary, or alluvial, (the collective extent of volcanic or
basaltic islands is exceedingly small,) the average density of

DENSITY OF 1 HE EARTH. 83

the superficies of that part of the outer crust of the globe
which is not covered by water, probably scarcely amounts to
between 2*4 and 2'6. If, with Rigaud, we take the ratio of
the dry land to the water-covered surface as 10 : 27, and
remember that ocean soundings have given a depth or stratum
of water of more than 27000 English feet/ it will follow
that the mean density of the external portion of our planet,
consisting partly of land and partly and more extensively of
water, scarcely attains the density of 1*5. It is certainly in
correct, as a celebrated geometrician, Plana, has remarked,
that the author of the " Mecanique celeste" should have
ascribed to the exterior terrestrial crust the density of granite,
which Density has itself been rated rather highly by Laplace
at =3*0, (28) whence he obtained the density of the centre
of the earth= 10-047. The latter is, according to Plana/
16-27 if we take the upper strata at = l'83 (which differs
little from 1*5 or 1'6 as the density of the total crust).'
The horizontal pendulum (the torsion balance) may, indeed, ;
have been called, as well as the vertical pendulum, a geo-
gnostical instrument ; but the geology of the inaccessible
internal parts of the earth is, like the astrognosy of " dark
heavenly bodies," only to be treated with great caution. In
the volcanic section of this work, I shall have to touch
on questions which have been propounded by others/
respecting currents in the generally fluid interior of the
earth, the probability or improbability of a movement of
periodical ebb and flood in detached basins not entirely
filled, and on the existence of spaces of less density beneath
upheaved mountain-chains. (29) We ought not to omit in
the " Cosmos" any consideration to which actual observa-
tions, or not remote analogies, appear to lead.
VOL. IY. D

84 INTERNAL HEAT OF THE EARTH,

/9. Internal Heat of the Earth, and its Distribution.

(Enlargement of the Representation of Nature in Kosmos,
Bd. i. S. 179— 184, 425—427, Anm. 7—30; Engl.
Vol. i. p. 161—167 and p. 405—407 ; Notes 137—140).

Considerations respecting the internal heat of the earth, the
importance of which has been so much enhanced by their
now so generally acknowledged connection with volcanic
action and phsenomena of elevation, are based partly on
direct, and therefore incontestable observation of tempera-
tures in spring?, borings, and mines ; and partly on analytical
combinations respecting the gradual cooling down of our
planet, and the influence which the diminution of tempera-
ture may have exercised on its velocity of rotation,(30) and
on the direction of internal thermal currents in primeval
time. The form of the flattened terrestrial spheroid is itself
again dependent on the law of increasing density in con-
centric superimposed non-homogeneous shells. The first-
mentioned, which is the experimental, and therefore more
assured portion of the investigation to which we here confine
ourselves, can, however, only shed light on the earth's
crust of inconsiderable thickness ; while the second, which
is the mathematical part of the inquiry, is, from the nature
of its applications, suited to afford negative rather than
positive results. Offering the charm which is afforded by
sagacious combinations of thought, (31) it leads to problems
which, together with conjectures respecting the origin of
volcanic forces, and the reaction of the molten interior
against the solid outer shell, or crust, cannot remain altogether
unnoticed. Plato's geognostic myth of the Pyriphlege-
thon,(32) as the origin or fount of all hot springs and volcanic

AND ITS DISTRIBUTION. 35

fiery currents, had its source in the early and generally felt
desire of discovering a common cause for a great and com-
plicated series of phsenomena.

Yiewing the variety of conditions which the surface of
the earth presents in respect to the reception and absorption
of the solar rays, and to the facility with which heat radiates
from the eartli into space, as well as the great diversity in
the conduction of heat by reason of the various composition
and density of different rocks, it is not a little wonderful
that where observations have been made with care and under
favourable circumstances, such (generally speaking) accordant
results for the increase of temperature with increasing depth
should have been obtained in such different localities. Borings,
especially where they are filled with water rendered rather
thick and muddy by the presence of clay, and thus less
favourable to the generation of internal currents, and sup-
posing there to be few lateral openings from whence affluents
from cross-fissures at different heights could be received,
present the greatest certainty ; and on this account, as well
as on account of their great absolute depth, we will b^gin
with the two most remarkable Artesian Wells, — that of
Grenelle at Paris, and of Neu-Salz\verk in the Oeynhausen
Soolbade, near Minden. I subjoin the most accurate deter-
minations at both.

According to the measurements of Walferdin,(33) to whose
ingenuity we are indebted for a variety of delicate apparatus
for determining temperatures at depths either of the sea or
of wells, the floor of the Abattoir du Puits de Grenelle is
36m,24 above the, sea. The highest level to which the
water of the spring ascends is 33m,33 higher. This total
height to which the water rises above the level of the sea

n 2

36 INTERNAL HEAT OF THE EARTH,

(69m,57) is in absolute elevation about 60 metres lower
than the out-cropping of the green-sand stratum in the
hills near Lusigny, south-east of Paris, to the infiltrations
of which the ascent of the water in the Artesian "Well of
Grenelle is ascribed. The water was reached by the boring
at a depth of 547m (1683 Paris, or 1794 English, feet) below
the floor of the Abattoir, or 1675 English feet below the level
of the sea : the whole ascent of the water, therefore, is
1903 English feet. The temperature of the spring is
27°'75 Cent. (81*95 Fahr.) The increase of temperature
is in this case 1° of the Centigrade thermometer for every
99 k Paris feet, or 1° of Fahr. for 58'9 English feet.

The height at which the boring commenced at Neu-
Salzwerk, near Eehme, is 218 Paris feet above the level of
the sea (i. e. above the Pegel at Amsterdam), and it reached
an absolute depth of 2144 Paris feet below the point at
which the work began. The Sool-spring, therefore, which
comes out impregnated with much carbonic acid, is 1926
French, or 2053 English, feet below the level of the sea ; a re-
lative depth the greatest perhaps which has ever been reached
by man. Its temperature is 32°'8 Cent., or 91°'04 Fahr.;
and as the mean annual temperature of the air at Neu-
Salzwerk is about 9°'6Cent., or 49°'28 Fahr., we may infer
an increase of temperature of 1° Cent, for 92'4 Paris, or 98'5
English, feet, or of 1° Fahr. for 54'72 English feet.

The boring of the well of Neu-Salzwerk,(34) compared
with that of Grenelle, has a greater absolute depth of 461
French feet ; a greater relative depth below the level of the
sea of 354 French, or 377 English, feet,; and the tempera-
ture of its water is 5°'l Cent., or 9°'18 Fahr., higher.
The increase of temperature is therefore nearly T'T

AND ITS DISTRIBUTION. 37

less rapid at Paris, or at the rate of about 7*1 French
feet more for each centesimal degree. I have already
noticed (35) that a very similar result was obtained by
Auguste de la Eive and Marcet in a boring examined by
them at Bregny, near Geneva, having a depth of only 680
French, or 724*7 English feet, although situated at more
than 1600 English feet above the Mediterranean.

If to the three wells which have been mentioned, which
attain depths of from 680 to 2144 Paris, or 725 to 2285
English, feet, we add that of Wearmouth, near Newcastle
(being the pit waters of the coal-mine in which, according to
Phillips, men work at a depth of 1404 French, or 1496
English, feet below the level of the sea), we find the remark-
able result, that at four places so distant from each other,
the increase of temperature only varies between 91 and 99
Paris feet for 1° Centigrade. (36) Such an accordance
cannot, however, from the nature of the means employed,
be everywhere looked for in inquiries into the internal heat
of the earth at given depths. Admitting that the meteoric
water falling on heights, and penetrating by infiltration,
produces by hydrostatic pressure, as in connected tubes, the
ascent of springs at lower points ; and that the subterranean
waters assume the temperature of the strata with which they
come into contact ; they may nevertheless, in certain cases,
have channels of communication with still deeper clefts,
from whence they may receive accessions of still warmer
water from unknown depths. Such an influence, which is
to be altogether distinguished from different degrees of con-
ducting power in the surrounding rocks, may take place at
points very distant from the shaft or boring of the well. It
is probable that the subterranean waters exist sometimes in

88 INTERNAL HEAT OF THE EARTH,

confined spaces, as in river-like fissures (and hence it often
happens that, of several borings in near proximity to each
other, only some are successful), while sometimes they may
form basins of wide extent in a horizontal direction, — a dis-
tribution very favourable for yielding supplies of water, and
which only in very rare cases betrays an open communication
with the surface of the earth by the presence of eels, shells,
and vegetable remains. If from such causes as have been
mentioned, ascending springs are sometimes warmer than
might be anticipated from the small depth of the boring
through which they rise, on the other hand an opposite
effect is produced by colder waters flowing in laterally through
openings from cross -fissures.

It has been remarked already that the maxima and minima
of the variations of atmospheric temperature, occasioned by
the height of the sun at the different hours of the day and
seasons of the year, reach points situated in the same vertical
line, at small depths below the surface, at very different
times. According to the always very exact observations of
Quetelet,(37) the diurnal variations cease to be sensible at
Brussels at the small depth of 3 '8 French feet ; and in respect
to annual variations, a thermometer placed 24 French, or
25 '6 English, feet below the surface at Brussels, did not show
the maximum of temperature during the year until the 10th
of December, and the minimum until the 15th of June.
Also, in the fine series of experiments made by Forbes in the
\icinity of Edinburgh, on the conducting power of different
kinds of rock, the maximum temperature in the basaltic trap
of Calton Hill was observed to take place on the 8th of
January, at a depth of 23 feet.(38) According to Arago's
series of many years' observations in the garden of the

AND ITS DISTRIBUTION. 39

Observatory at Paris, only very slight differences of tempe-
rature are discernible at 28 Paris, or 30 English, feet below
the surface. Bravais still found a difference of 1° Cent.
(l°-8 Pahr.) at a depth of 26i Prench feet in the high North,
at Bossekop in Pinmarken, lat. 69° 58'. The difference be-
tween the maximum and minimum temperatures of the year
becomes rapidly less as we descend; diminishing, according
to Pourier, in geometrical, as the depth increases in arith-
metical, progression.

In reality, however, the depth of the " invariable stra-
tum" below the surface, or the depth at which no sensible
change of temperature takes place, is not everywhere the
same. It depends on the latitude of the place, on the heat-
conducting power of the rock, and on the amount of
difference between the temperatures of the hottest and coldest
seasons. In the latitude of Paris (48° 50'), the observa-
tions in the " Caves de 1'Observatoire " give 86 Paris feet,
or 92 English feet as the depth, and 11°'834 Cent., or
53°'3 Pahr., as the temperature of the "invariable stratum."
Since Cassini and Legentil in 1783 fixed a very exact
mercurial thermometer in those subterranean spaces which
are parts of ancient stone- quarries, the mercury has risen
0°'22 (Cent.) in the tube.(39) It is still doubtful whether
this rise is caused by an accidental alteration in the thermo-
meter scale, (which, however, was very carefully verified by
Arago in 1817), or whether it is to be ascribed to a real
increase of temperature. The mean temperature of the air
at Paris is 10°'822 Cent., or 51°*4<8 Pahr. Bravais believes
the thermometer in the Caves de 1'Observatoire to be below
the invariable stratum ; although Cassini considered he had
found a difference of two-hundredths of a centesimal degree

-1*0 INTERNAL HEAT OF THE EARTH,

between the winter and summer temperatures, (40) the winter
temperature being the higher. If we take the mean of many
observations of temperature at and below the surface of the
ground between the parallels of Zurich (lat. 47° 22') and
Upsala (lat. 59° 51'), we find a mean increase of 1° Cent, for
the depth of 67i French, or 72 English, feet. The diffe-
rences due to latitude are not more than from about%13 to 16
English feet, and do not show any regular systematic varia-
tion from south to north ; probably because the undoubtedly
existing effect of latitude is at present inextricably blended
with the influence of the varying heat-conducting powers of
the different rocks, and with errors of observation.

As, according to the theory of the distribution of heat, the
underground stratum in which we begin to find no difference
of temperature in the course of the annual cycle, is so much
the less deep beneath the surface as the maxima and minima
of temperature in the year are less distant from each other,
my friend Boussingault was led by this consideration to the
ingenious and convenient method of determining the mean
temperature of places within the tropics (more particularly
within 10° on either side of the equator) by observing a
thermometer buried about a foot deep within a well-covered
space. At the most different hours, and even in different
months, (as is shown by the experiments of Colonel Hall
near the sea-shore of Choco, in Turnaco ; those of Salaza at
Quito ; and those of Boussingault in the Yega de Zupia, at
Marmato, and Anserma Nuevo in the Cauca Valley), the
temperature did not vary two-tenths of a centesimal degree
(0°'36 Fahr.), and was identical within almost the same
limits with the mean atmospheric temperature at places
where the latter had been derived from hourly observations.

AND ITS DISTRIBUTION, 41

Tt is well deserving of notice that this agreement remained
the same whether the " thermometric soundings" of only
one foot were made on the hot shores of the Pacific at
Guayaquil and Payta, or at an Indian village on the side of
the Volcano of Purace, the height of which above the level
of the sea I found by barometric measurements to be 1356
toises (or 8671 English feet). The mean temperatures
differed at these different elevations fully 14° Cent. (25°'2
Eahr.) (41)

Two observations made by myselt in the mountains of
Peru and Mexico, are, I think, deserving of particular atten-
tion, because made in mines situated at elevations higher
than the summit of the Peak of Teneriffe ; and higher, I
believe, than any mines into which a thermometer has yet
been taken. Thirteen thousand feet above the level of the
sea, I found the subterranean air 14° Cent, warmer than
the external air. The little town of Micuipampa,(42) in
Peru, is situated, according to my determinations of its
geographical position and elevation, in 6° 43' S. latitude,
and at a height of 1857 toises (11875 English feet) above
the sea, at the foot of the Cerro de Gualgayoc, celebrated for
its mineral (silver) wealth. The summit of this almost iso-
lated, castle-like, and picturesque mountain is 240 toises
above the pavement of the street of Micuipampa. The
temperature of the external air at a proper distance from
the entrance of the Mina del Purgatorio was 5°'7 Cent.
(42°*26 Eahr.) ; but everywhere in the interior (at a "height
of about 2057 toises, or 15153 English feet above the sea) 1
saw the thermometer show 19°*8, being a difference of
14°-1 Cent., or 250>4 Eahr. The limestone rock was
perfectly dry, and there were very few miners at work. In

42 INTERNAL HEAT OF THE EARTH,

the Mina de Guadalupe, which is at the same elevation, I
found the internal temperature 14°'4 Cent. (57°'9 Fahr.),
the difference from the external air being therefore 8°*7, or
15°* 8 Fahr. The waters which were streaming down in this
very wet mine showed 11°'3 Cent. (52°'3 Fahr.) The
mean annual temperature of the air at Micuipampa is
probably not above 1\° Cent., or 45^° Fahr. In Mexico,
in the rich silver mines of Guanaxuato, I found at the
Mina de Yalenciana(43) the temperature of the external air
near Tiro Nuevo (7122 Paris, or 7590 English, feet above
the sea) 21°'2 Cent., or 70°'16 Fahr.j and the air in the
deepest part of the mine, in the Planes de San Bernardo
(1630 English feet below the opening of the shaft of Tiro
Nuevo), fully 27° Cent., or 80°'5 Fahr., which is about
the mean temperature of the shore of the Gulf of Mexico.
In a part of the mine situated 147 feet higher than the floor
of the Planes de San Bernardo, there bursts out from the
rock a spring of water of the temperature of 29°'3 Cent.,
or S4°'74 Fahr. The mountain town of Guanaxuato,
situated, according to my determination, in 21° N. lat.,
has a mean temperature falling, approximately, somewhere
between 15°-8 and 16°'2 Cent., (60°'44 and 61°-16 Fahr.}
It would be unsuitable to enter here into conjectures, diffi-
cult to establish on any very certain grounds, as to the
causes, possibly of very local influence, which thus raise
the temperature of subterranean spaces, in ranges of moun-
tains from six to thirteen thousand feet high.

A remarkable contrast to the above is presented by
the circumstances of the ground-ice in the steppes of
Northern Asia. Even the existence of this phenomenon
was formerly doubted, notwithstanding the testimony early

AND ITS DISTRIBUTION. 43

given by Gmelin and Pallas concerning it. It is only very
recently, that, through the excellent investigations of Erman,
Baer, and Middendorff, correct views have been gained
respecting the extent and thickness of this stratum of sub-
terranean ice. Prom the descriptions of Greenland by
Crantz, of Spitzbergen by Martens and Phipps, and of the
shores of the Sea of Kara by Sujeff, an incautious genera-
lisation had represented the whole northern part of Siberia
as destitute of vegetation, and with a constantly frozen and
snow-covered surface, even in the plains. The extreme limit
of the growth of high forest-trees in Northern Asia is not
the parallel of 67° latitude, as was long assumed, and as
is actually the case near Obdorsk, owing to the influence of
sea-winds and the vicinity of the Gulf of Obi. The valley of
the great river Lena has lofty trees up to the latitude of
71°. In the desert islands of New Siberia large herds of
reindeer and countless lemmings still find sufficient vegetable
sustenance (44). The two Siberian journeys of Middendorff,
a traveller eminent for the spirit of observation as well as for
boldness of enterprise and perseverance in laborious re-
search, were made from 1843 to 1846, and were directed
northwards, into the Taymir country, to nearly 76° N. lat. ;
and to the south-east as far as the upper Amour and the
Sea of Ochozk. The first of these adventurous journeys led
the learned traveller into a previously wholly unvisited region,
the more important and interesting because situated at an
equal distance from the east and west coasts of the old
continent. The instructions of the Petersburg Academy
of Sciences, commended to him as a leading object of his
expedition (in canjunction with the limits of the various
forms of organic life towards the high north, depending

44 INTERNAL HEAT OF THE EARTH,

principally on climatic relations), the accurate determina-
tion of the temperature of the soil, and the thickness of the
subterranean Or ground-ice. He accordingly examined the
latter questions by means of borings and excavations from
21 to 61 feet deep, at more than twelve points (at Turu-
chansk, on the lenisei, and on the Lena), at distances of
from 1600 to 2000 geographical miles apart.

The most important point, however, for geothermic obser-
vations, was the Schergin Shaft(45) at lakutzk (lat. 62° 2'),
where the subterranean icy stratum has been pierced for a
thickness of more than 382 English feet. Thermometers
have been inserted in the side walls of the shaft, at
eleven points situated vertically in respect to each other,
between the surface of the earth and the deepest part of
the shaft reached in 1837. The observer, in descend-
ing, had to read the thermometer scales standing in a
bucket, and holding by one hand to the rope. The series
of observations, of which the mean error is estimated at only
0°'25 Cent., or 0°*45 Pahr., extended over the interval
from April 1844 to June 1846. The decrease of cold was
not indeed always in proportion to the depth at each of the
separate points, but on the average the increase of tempera-
ture with increasing depth was found as follows : —

Engrl. Feet. Reaumur. Fahrenheit,

o o

50 .... -6-61 .... 17-13

100 .... -5-22 .... 20-25

150 .... -4-64 .... 21-56

200 .... -3-88 .... 23-27

250 .... -3-34 .... 24-49

382 . . -2-40 26-60

AND ITS DISTRIBUTION. 45

By a thorough discussion of all the observations, Mid-
dendorff obtained, as the general rate of increase of
temperature, (48) 1° of Reaumur for between 100 and
117 English feet, or 1° Pahr. for from 44-4 to 52'1
English feet. This is a more rapid rate of increase than
is given by several very accordant results from different
excavations in Middle Europe (see above, p. 37). The
mean annual temperature of lakutzk is considered to
be -8°13 Reaum., or 13°'71 Pahr. Neveroff's observa-
tions, continued during fifteen years (1829 — 1844), give
the variation of temperature between winter and summer
so great, that sometimes the temperature of the air, for
14} successive days in July and August, is from 20° to
23°-4 Reaum., or from 77° to 84°-6 Eahr. ; while for
120 successive days in winter (November to February)
the cold fluctuates between —33° and — 44°*8 Reaum.,
or — 42°-25 and — 6S°'S Fahr. The depth beneath
the surface of the earth at which the lower limit of the
frozen ground, or the temperature of 0° Reaum. and Cent.,,
or 32° Fahr., would be met with, is calculated approxi-
mately from the increase of temperature found in piercing
the icy stratum so far as lias yet been done. Middendorff's
estimate, from his observations in the Schergin Shaft, in
accordance with the much earlier one of Erman, assigns from.,
612 to 642 French, or 652 to 684 English, feet. On the
other hand, the increase of temperature in the excavations
of Mangan, Schiloff, and Davydoff (scarcely four miles
from lakutzk, in the chain of hills on the left bank of
the Lena), which, however, are not quite sixty feet deep,
would place the temperature of 0° (or 32° Fahr.), at a
depth of little more than 300 feet.(4?) Is this great

46 INTERNAL HEAT OF THE EARTH,

difference in the two situations more than accidental?
A numerical determination depending on such inconsiderable
depths must be regarded as exceedingly uncertain, and the
increase of temperature may not always follow a uniform
law. Ought we to infer that if a gallery were to be driven
horizontally for many hundred fathoms from the lowest
depth of the Schergin Shaft, it would encounter everywhere,
and in every direction, frozen earth, and that with a tempe-
rature of — 2°-5 Cent, below 32°?

Schrenk has examined the ground-ice in 67i° lat. in the
Samoied country, at Pustoienskoy Gorodok. In this case
the sinking of the well was aided by the application of fire.
In the middle of summer the frozen stratum was encountered
at a depth of little more than 5 feet ; it was followed through
a thickness of 67 feet, when the work was suddenly stopped.
The neighbouring lake of Ustie continued to be frozen over
and passable by sledges throughout the summer of 1813.(48)
On my Siberian expedition with Ehrenberg and Gustav
Hose, we had a hole dug in the boggy or turfy soil near
Bogoslowzk (lat. 5 9° 44') on the road to the Turjin Mine(49)
in the Oural Mountains. At a depth of about 5J feet, we
came upon pieces of ice forming a kind of breccia with
frozen earth ; then solid ice, which continued to the depth
of 10 or 11 feet further, at which we left off.

The geographical extent of the ground- or subterranean -
ice, — i. e. the position and course of the southern limit, in
the old continent from Scandinavia to the eastern coast of
Asia, at which we begin to find in August, and throughout
the year, ice or frozen earth at a certain depth, is, according
to MiddendorfFs sagacious generalisations from observation,
like all other geothermic relations, often more <lq>en-

AND ITS DISTPaBUTION. 47

dent on local circumstances than on the temperature of the
atmosphere. No doubt the latter has on the whole the most
determining influence; but yet, as Kupffer has already
remarked, the isogeothermal lines are not parallel in their
concave and convex inflections to the climatic isothermals
which are given by the mean atmospheric temperature. The
amount of rain, and the depth to which it penetrates, the
ascent of warm springs from depths, and the very various
conducting power of the soil,(50) all appear to be particularly
influential. " At the northernmost point of Europe, in Pin-
mark en, in 70° and 71°lat, there is no connected 'ground-
ice/ Proceeding eastward to the valley of the Obi,
we find the ground-ice at Obdorsk and Beresow 5° more
southerly than the North Cape. The cold of the ground
continues to increase as we advance eastward, with the ex-
ception of Tobolsk on the Irtish, where the temperature of
the ground is lower than at Witimsk in the valley of the
Lena, which is 1° more to the north. At Turuchansk
(65° 54' lat.) the ground is still not frozen, but the limit
at which the ground-ice begins is very near. At Amginsk,
south-east of lakutzk, the ground is as cold as at Obdorsk,
5° further to the north ; so also at Oleminsk on the
Tenisei. Prom the Obi to the lenisei, the curve which
bounds the ground-ice appears to rise 2° of latitude more
towards the north; and then to re-descend, to cross the
valley of the Lena, almost 8° south of the parallel in which
it crosses the lenisei. Further to the east the line re-ascends
towards the north." (51) Kupffer, who has visited the mines
of Nertschinsk, has pointed out, that apart from the great
connected region of the ground-ice, the same phenomenon
occurs more to the south in detached patches, or as it were.

48 INTEKXAL HEAT OF THE EARTH.

in an insular manner. It is quite independent, generally
speaking, of the limits of vegetation and of the presence of
high trees.

The gradual acquirement of a true general or cosmical
view of the thermal relations of the crust of the earth in
the northern parts of the old continent, is an important
step in the progress of knowledge, as is the recognition that
the limit of the ground-ice, as well as the limits of particular
mean annual temperatures, and of the growth of trees, are
found in very different latitudes in different meridians, —a
fact which must occasion the production of perpetual thermal
currents in the interior of the earth. In the north-west
of America, Franklin found the ground frozen in the
middle of August, at a depth of 1 7 inches. At a more
easterly part of the coast, in 71° 12' latitude, Eichardson
saw in July the ice-stratum thawed as far down as three feet
below the herb-covered surface. It is to be desired that
scientific travellers may soon obtain for us a more general
knowledge of the geothermic relations subsisting in this part
of the world and in the southern hemisphere. Eesearch
into the connection of phenomena leads us most surely to
the recognition of the causes of apparently complex or
anomalous facts, or of what is sometimes too hastily called
irregulanty.

TERKESTRIAL MAGNETISM. 49

y. Magnetic Activity of the Earth in its three
manifestations of Force, Inclination, and Declina-
tion— Points (termed Magnetic Poles] at which the
Inclination is 90° — Line on which no Inclination is
observed (Magnetic Equator] — Four Points of greatest,
but unequal, Intensity of the Magnetic Force — Curve
or Line of the weakest Magnetic Force — Extraor-
dinary Disturbances of the Declination (Magnetic
Storms] — Polar Light.

(Extension of the Eepresentation of Nature, in Kosmos,
Ed. i. S. 184—208, and 427—442, Anm. 11—49 ;
Bd. ii. S. 372—376, and 515, Anm. 69—74; Bd. iii.
S. 399—401, and 419, Anm. 30 : Engl. Yol. i. p.
167—189, and 407—424, Notes 141—179; Vol. ii.
p. 331—335, and cxix., Notes 509—514, bis; Yol. iii.
p. 289—290, and civ., Note 489.)

The magnetic constitution of our planet can only be inferred
from those manifestations of its magnetic force which pre-
sent measurable relations in reference either to place or
time. These manifestations are characterised \yy perpetual
variability in the phsenomena which they present, and this
in a far higher degree than such other variable phenomena
as the temperature, aqueous vapour, and electric tension of
the lower strata of the atmosphere. The continual changes
which take place in the kindred magnetic and electric con-
ditions of matter, form an essential distinction between the
phenomena of electro-magnetism, and those which depend
on that primary, fundamental force of molecular and mass-
attraction of all matter, which, at unchanged distances,
VOL. TV. E

50 TERRESTRIAL MAGNETISM.

remains ever the same. The investigation of the dominion
of law amidst variability, is itself the proximate object, or
the goal immediately aimed at, in every investigation into a
force in nature. "While the labours of Coulomb and Arago
have demonstrated that the electro -magnetic process can be
elicited in the most various substances, Faraday's brilliant
discovery of diamagnetisin, on the other hand, shows us
here also, in the distinction of north and south axiality in
one class of substances, and east and west axiality in
others, that influence of the diversity or heterogeneity of
substances of which molecular or mass-attraction is wholly
independent. Oxygen gas, enclosed in a thin glass tube,
when placed under the action of a magnet in its vicinity,
assumes " paramagnetically" a north and south direction,
like iron; nitrogen, hydrogen, and carbonic acid gas,
remain unaffected ; while phosphorus, leather, and wood,
under the same circumstances, range themselves "dia-
magnetically," i. e. in an equatorial or east and west
direction.

The facts known in Greek and Roman antiquity were :
the adhesion of iron to the loadstone ; magnetic attraction
and repulsion ; the propagation of the attracting influence
through iron vessels, and also through rings (52) forming
links in a chain (one ring being in contact with the load-
stone) ; and the non-attraction of wood, or of other metals
than iron. Of the polar directive force which magnetism
could impart to a body susceptible of its influence, the
early western nations (Phoenicians, Etruscans, Greeks and
Romans) knew nothing. It is not until the llth and 12th
centuries of our era that we find among the European
nations any knowledge existing of this " directive force,"

TE11EESTRIAL MAGNETISM. 51

which has exerted so powerful an influence on the improve-
ment and extension of navigation, and, in consequence of
the practical and material services which it has thus rendered,
has given so persistent an impulse to the study of an all-
pervading, yet formerly little regarded, natural force. In
the history of the principal epochs in the Physical Contem-
plation of the Universe, (53) the subject of the earth's
magnetism, which we now bring under one general view,
assigning the several sources or authorities from whence the
information is derived, fell under several different chrono-
logical heads, and was there divided among different
sections.

It is among the Chinese that we find the first application
of the magnetic directive force to practical purposes by the
employment of magnetised needles floating on water, which
goes back to an epoch anterior perhaps to that of the
Doric migration and the return of the Heraclides to the
Peloponnesus. It is remarkable that in this part of Asia
the use of needles, the south end of which was the one dis-
tinguished, as with us the north, commenced in land journeys
before their employment in sea navigation. Instead of a
ship's compass they used a magnetic car, on the front part
of which a floating needle carried a little figure, whose out-
stretched arm and hand pointed to the south. Such an
apparatus, Pse-nan (indicator of the south), was presented,
under the dynasty of the Tscheu, 1100 years before our era,
to the ambassadors from Tunkin and Cochin-china, to
guide them across the great plains on their return from
China to their own country. The magnetic car continued
to be used as late as the 15th century. (54) Many such
instruments were kept in the Emperor's palace, and it

E 2

52 TERRESTRIAL MAGNETISM".

was also customary to use them in determining the direction
of the walls when building Budhistic convents. Their
frequent employment led some of the more acute thinkers to
form physical speculations respecting the nature of the
magnetic phenomena. The Chinese panegyrist of the
magnetic needle, Kuopho, a contemporary of Constantine
the Great, compared the attractive force of the magnet
to that of rubbed amber. " It is/' he says, " as if a
mysterious breath passed through both, communicating
itself with the swiftness of an arrow." The symbolical
expression of "a breath" reminds us of the equally
symbolical expressions used in Grecian antiquity by
the founder of the Ionic school, Thales, in reference
both to the magnetic stone and> to amber, which were
said to be " animated," or imbued with a " soul" or " spirit,"
meaning evidently an indwelling principle of motive
activity. (55)

As the too great mobility of these Chinese floating
needles was found inconvenient, they were replaced in
the beginning of the 12th century by a different con-
struction, in which the needle was suspended freely in
the air by a cotton or silken thread, quite in the manner
of the " Coulomb" suspension, which was first employed in
Europe by Gilbert. With this improved apparatus, (56)
the Chinese even determined early in the 12th century
the amount of the west declination, which appears to
undergo only very slow and minute changes in that part
of Asia. At length the compass came to be employed at
sea as well as on land. Under the dynasty of Tsin, in the
4th century, Chinese ships guided by compasses visited the
ports of India and the eastern coast of Africa.

TERRESTRIAL MAGNETISM. 53

Two centuries earlier, in the reign of Marcus Aurelius
Antoninus (called An-tun by the Chinese writers of the
dynasty of Han), Roman legates had come by water by Tunkin
to China ; but it was not through so transitory a communi-
cation that the use of the compass reached Europe ; where
it was not introduced until the 12th century, after its use
had become general throughout the Indian seas and the coasts
of Persia and Arabia. The introduction was effected either
directly by the influence of the Arabians, or through the me-
dium of the Crusaders, who since 1096 had been in contact
with Egypt and the Levant. In historical inquiries of this
class, the only epoch which can be assigned with certainty is
that which must be regarded as the latest or limiting date.
In a political satirical poern of Guyot of Provins, in 1199, the
mariner's compass is spoken of as an instrument long known
among the nations of Christendom ; and this is also the
case in a description of Palestine which we owe to Jaques
de Yitry, Bishop of Ptolemais, and which was finished
between 1204 and 1215. Guided by the compass, the
Catalans sailed to the northern isles of Scotland, and to
the west coast of tropical Africa ; the Basques to the whale
fishery ; and the Normans to the Azores (the Bracix Islands
of Picigario). In the first half of the 13th century,
the Spanish " Leyes de las Partidas," the work " del sabio
Hey Don Alonso el Nono," extols the compass-needle as the
faithful mediatrix (medianera) between the "magnetic
stone"— "la piedra"— and the "north star." Gilbert, in
his celebrated work "de Magnete Physiologia nova/'
speaks of the compass as a Chinese invention, but adds,
which is clearly incorrect, that it was first brought to Italy by
Marco Polo, "qui apud Chinas artem pyxidis didicit." As

54 TERRESTRIAL MAGNETISM.

Marco Polo began his travels in 127 1, and returned in
] 295, the evidence of Guyot de Prcvins and Jaques de
Yitry shows that the compass was used in European naviga-
tion at least sixty or seventy years before the commencement
of his travels. The names of Aphron and Zohron, given to
the north and south poles of the needle by Vicentius of
Beauvais in his "Mirror of Nature" (1254), indicate that
Arab pilots were the channel through which the nations of
Europe received the Chinese compass, being obviously
derived from that learned, ingenious, and active race,
traces of whose language are so frequent in our star-maps,
though appearing there too often in a mutilated form.

Prom all these circumstances, there can be no doubt that
the general employment of the magnetic needle in ocean
navigation by the Europeans from the 12th century, (and
in a limited degree probably still earlier), proceeded from
the Mediterranean and its shores, and took place chiefly
through the agency of Moors, Genoese, Venetians, Major-
cans, and Catalans. Catalan sailors, under the conduct of
the celebrated navigator, Don Jaime Perrer, advanced in
1346 to the mouth of the Bio de Ouro, 23° 40' N. latitude,
on the west coast of Africa; and from the evidence of
Eaymund Lully (in his nautical work, Penix de las Mara-
villas del Orbe, 1286), it appears that long before Jaime
Perrer, the Barcelonese used sea-charts, astrolabes, and sea-
compasses.

The knowledge of the existence of a greater or less
amount of magnetic declination (its early name was simply
" variation," without any adjunct) had naturally been
also diffused over the Mediterranean, by report from China,
through the medium of Indian, Malay, and Arab mariners.

TERRESTRIAL MAGNETISM. 55

This element, so indispensable for the correction of ships'
reckonings, was, at that early period, determined less often
by sunrise and sunset than by the pole-star, and in either
case with much uncertainty ; it was, however, entered on
nautical charts, — for example, on the rare chart of Andrea
Bianco, sketched in 1436. Columbus, who was certainly
not the first who recognised the fact of magnetic decli-
nation,— any more than Sebastian Cabot, of whom it has
sometimes been stated, — has the praise of having been the
first who determined the position of a line of no declination,
which he did astronomically, in 2j degrees east of the Island
of Corvo (one of the Azores), on the 13th of September,
1492. In traversing the western part of the Atlantic
Ocean, he was the first who observed the "variation"
change gradually from north-east to north-west. He was
led thereby to conceive the idea, which has so often engaged
the attention of navigators in later centuries, of making use
of the declination lines, — which he imagined to be parallel
to the meridians, — for finding the- longitude. We learn
from his journals, that, on his second voyage (1496), when
uncertain of his position, he sought it by means of declina-
tion observations. A view of the possibility of such a
method was also, without doubt, that infallible secret of the
longitude at sea, which, on his death-bed, Sebastian Cabot
boasted of possessing through special divine revelation.

In the excitable imagination of Columbus, there were
connected with the Atlantic line of no declination other
and rather fanciful ideas of supposed change of climate,
anomalous figure of the terrestrial spheroid, and extraor;
dinary views respecting the movements of the heavenly
bodies, in all of which he found reasons for proposing the

56 TERRESTRIAL MAGNETISM.

conversion of a physical line into a political line of demar-
cation. The "raya," on which the "agujas de marear"
point direct to the pole-star, became the boundary line
between the crowns of Castile and Portugal by virtue of a
Papal decree, whose arrogance proved of great and lasting,
though undesigned and unforeseen, benefit to the enlarge-
ment of nautical astronomy and the improvement of mag-
netic instruments, in consequence of the importance which
naturally attached to determining, with astronomical accuracy,
the geographical longitude of such a boundary on both sides
of the equator. (Humboldt, Examen critique de la Geogr.
, T. iii. p. 54.) Felipe Guillen, of Seville (1525), and, at a
still earlier period, probably the cosmographer Alonso de
Santa Cruz, teacher of mathematics to the young Emperor
Charles V., constructed new " variation compasses/' with
which solar altitudes could be taken. In 1530 — a century
and a half, therefore, before Halley — the same Alonso
de Santa Cruz drew the first general "variation map/'
founded, it must be admitted, on very imperfect materials.
The voyage of Juan Jayme, who sailed from the Philippines
to Acapulco with Francisco Gali in 1585, solely for the
purpose of trying, during the long passage across the Pacific,
a new declination instrument which he had invented, shows
how animated an impulse had been given in the 16th
century, and after the death of Columbus, to the study of
terrestrial magnetism, by the controversy respecting the
Papal " line of demarcation."

This disposition to pursue observation could not but be
accompanied by its unfailing attendant, — if not, as is still
oftener the case, its precursor, — a fondness for theoretical
speculations. Many of the old sea-stories of the Indians,

TERRESTRIAL MAGNETISM. 57

as well as of the Arabs, speak of rocky islands. which cause
disasters to mariners, either by drawing out, by virtue of
their natural magnetic power, all the iron which held the
wooden framework of the ship together, or by attracting
and immovably enchaining it. Under the influence of
such imaginations, the idea of the polar conjunction of all
the lines of declination had associated with it, in early times,
the more material image of a high magnetic mountain, in
near proximity to one of the poles of the earth. In the
remarkable map of the new continent, appended to the
Roman edition of 1508 of Ptolemy's Geography, we find
to the north of Greenland (Gruentlant), which is repre-
sented as belonging to the eastern part of Asia, the north
magnetic pole figured as a mountainous island rising
out of the sea. Its position was gradually removed farther
to the south in the " Breve Compendio de la Sphera" of
Martin Cortez, in 1545, as well as in Livio Sanuto's
"Geographia de Tolomeo," in 1588. Great expectations
were attached to reaching this point, which was termed " el
calamitico •" and from some notion, which was very late
in disappearing, it was supposed that whoever should reach
the magnetic pole would find " alcun miraculoso stupendo
effetto."

Until near the end of the 16th century, the magnetic
declination, which is the element exercising the most direct
influence on the requirements of navigation, was the only
one which received attention. Instead of the one " line of
no variation," found by Columbus in 1492, the learned
Jesuit Acosta, in 1589, thought, from the information which
he had gathered from Portuguese seamen, that he could state,
in his excellent work, " Historia natural de las Indias," the

58 TERRESTRIAL MAGNETISM.

existence of four such lines. As a ship's reckoning requires
an exact knowledge of the distance passed over, as well as
of the exact direction of the course (given by the angle mea-
sured by the corrected compass), the introduction of the
use of the " log/' imperfect as this kind of measurement is
even at the present day, was yet an important epoch in the
history of navigation. I think I have proved, contrary to
the prevailing opinion hitherto, that the first certain evi-
dence (57) of the employment of the log (la cadena de la
popa, la corredera) is to be found in Antonio Pigafetta's
ship's journal in Magellan's voyage, in an entry appertaining
to the month of January, 1521. Columbus, Juan de la
Cosa, Sebastian Cabot, and Vasco de Gama, were all un-
acquainted with the log and its applications. They
estimated the ship's rate of movement by the eye alone,
and judged of the distance passed over by " hour glasses/'
i. e. by the running-out of sand in the " ampolletas." At
length, in 1576, in addition to the horizontal declination
from the geographical north, which had been so long ex-
clusively regarded, the second element, the inclination, or
dip of the needle below the horizontal line, came also to
be measured. Robert Norman, the first who determined
it, did so in London with an instrument of his own inven-
tion, and with no inconsiderable degree of accuracy. Fully
two centuries more elapsed before any attempt was made to
measure the third element, viz. the intensity of the earth's
magnetic force.

A man whom Galileo admired, although Bacon altogether
overlooked his merits, William Gilbert, brought forward,
at the end of the 16th century, the first enlarged and
comprehensive view(58) of the earth's magnetism. He first

TERRESTRIAL MAGNETISM. 59

distinguished clearly between magnetism and electricity in
respect to their effects, while he yet regarded both as
emanations of a single fundamental force inherent in all
matter. As is the privilege of genius, from feeble ana-
logies he successfully divined much. Prom the clear con-
ceptions which he formed of terrestrial magnetism (de
magno magnete tellure), he even at that time correctly
ascribed the origin of magnetic poles in the upright iron
bars of crosses on old church-towers, to impartation from
the magnetism of the earth. He was the first in Europe
who showed how to render iron magnetic by rubbing it with
the loadstone, which indeed the Chinese had known and
practised almost 500 years before.(59) Gilbert also already
gave steel the preference over soft iron, because capable of
appropriating to itself and retaining more permanently the
magnetic force imparted to it.

In the course of the 17th century, the navigation of
the Dutch, English, Spaniards, and French (which had so
greatly increased in extent through the improvement in the
means of determining the direction of a ship's course, and the
length or amount of distance traversed by her), augmented
the knowledge already possessed of different isogonic lines,
and more particularly of the lines of no declination, which,
as already mentioned, Acosta had attempted to form into a
system. (60) In 1616, Cornelius Schouten pointed out in
the middle of the Pacific Ocean, near the Marquesas, places
where the needle had no variation. We still find in this
region a singular " closed" system of isogonic lines, within
which the declination changes in amount in successive con-
centric curves. (61) The eagerness to discover new methods
of determining the longitude, for which it was thought not

60 .TERRESTRIAL MAGNETISM.

only that the declination, but also that the inclination, of
the magnetic needle might be made available, — (such a use
of the inclination, with a clouded starless sky, " per acre
caliginoso," Wright called " worth much gold,") (62) — led
to the construction of many and various magnetic instru-
ments, and stimulated the activity of observers. The Jesuit
Cabeus of Ferrara, Ridley, Lieutaud in 1668, and Henry
Bond in 1676, distinguished themselves in this way. The
controversy between Bond and Beckborrow, together with
Acosta's view of the existence of four lines of no declination
dividing the entire surface of the globe into four parts,
may, perhaps, have had some influence on Halley's theory
of " four magnetic poles, or points of convergence/' pro-
jected as early as 1683.

Halley's name constitutes an important epoch in the
history of terrestrial magnetism. He assumed the existence
in each hemisphere (northern and southern) of two mag-
netic poles, a stronger and a weaker pole ; and we now find
an analogous distribution of the magnetic force in four
points of maximum force, two in each hemisphere, one
stronger than the other, as shown by the rapidity of the
vibrations performed by a needle oscillating in the direction
of the magnetic meridian. The strongest of Halley's four
poles was placed in 70° S. lat. and 120° E. long, from
Greenwich, or almost in the meridian of King George's
Sound, in New Holland. (63) Halley's three voyages in
1698, 1699, and 1702, followed the first sketch of his theory,
which was founded only on the observations obtained on his
voyage to St. Helena seven years before, and on the
imperfect variation (declination) observations of Baffin,
Hudson, and Cornelius von Schouten. Halley's expeditions

TERRESTRIAL MAGNETISM. 61

were the first undertaken by a government for a great
scientific object, viz. for the investigation of an element
of the earth's force, on which the safety of navigation
especially depends. He advanced as far as 52° S. lat.,
and was thus enabled to construct the first extensive
variation- or declination-chart, which chart now supplies
to the theoretical investigators of the 19th century a point
of comparison, although not indeed a very remote one, for
the representation of the progressive change of position of
the declination lines.

It was a happy undertaking of Halley's to connect graphi-
cally by lines or curves all the points on the map where the
magnetic decimation was the same.(64) Clearness of repre-
sentation, and the advantage of gaining a general view of the
connection of detached results, were thus first introduced.
My isothermal lines, i. e. lines of equal temperature (mean
annual, summer, or winter temperature), which have been
favourably received by physicists, were formed in strict
analogy with Halley's isogonic curves. The object of the
isothermal lines, especially since their great extension and
improvement by Dove, has been to throw light on the
distribution of temperature over the earth's surface, and on
the dependence of that distribution in a great degree on the
configuration, extent, and relative position of the portions of
the surface occupied by land and water. Halley's purely
scientific expeditions stand out as the more remarkable,
because they were not designed, like so many subsequent
expeditions undertaken at the public expense, as voyages
for geographical discover yt but were strictly for scientific
research. Halley's stay at St. Helena in 1677 and 1678
had for its fruit, in addition to the data furnished to the

62 TERRESTRIAL MAGNETISM.

knowledge of terrestrial magnetism, an important catalogue
of southern stars, which it may be remarked, in passing, was
the first star- catalogue undertaken since the combination of
telescopes with measuring apparatus introduced by Morin
and Gascoigue.(65)

As the close of the 17th century had been marked by
progress towards a better knowledge of the position of the
declination lines, and by the first theoretical attempt to
determine their points of convergence as magnetic poles, so
the first half of the 18th century produced the discovery of
the horary periodic variation of the declination. Graham, in
London, in 1722, has the uncontested merit of being the
first to observe these variations with accuracy and perseve-
rance. Celsius and Hiorter, at Upsala, who were in episto-
lary communication with Graham, (66) further enlarged the
knowledge of the phsenomena in question. It was not until
the latter' part of the century, in 3784 — 1788, that Brug-
mauns and Coulomb, the latter gifted with a more mathe-
matical mind, penetrated more deeply into the essence
of terrestrial magnetism. Their acutely devised physical
experiments embraced the magnetic attraction of all matter,
the distribution of the force in a bar-magnet of a given
form, and the law of magnetic action at a distance. In the
methods adopted by them for obtaining exact results, they
sometimes employed the vibrations of a horizontal needle
suspended by a thread, and sometimes deflections measured
by a torsion balance.

Science is indebted for the first knowledge of the variation
in the intensity of the earth's magnetic force at different points
of its surface, obtained by the vibration of a needle suspended
vertically and placed in the magnetic meridian, solely to the

TERRESTRIAL MAGNETISM. 63

sagacity of the Chevalier Borda, not by successful experi-
ments made by himself personally, but by those suggested
by his sagacious anticipations, and carried into execution in
consequence of the influence which he perseveringly exercised
on travellers and voyagers preparing for distant expeditions.
His long-cherished conjectures were first confirmed by
Lamanon, the companion of La Pe rouse, in the years
1785 — 1787. These observations, although their results
had been made known so early as the summer of the last-
named year to the Secretary of the Academic des Sciences,
Condorcet, remained unnoticed and unpublished. The
credit of the first, although on this account incomplete,
recognition of the important law of the variation of the mag-
netic force with the magnetic latitude, belongs without
dispute to the ill-fated expedition of La Perouse,(67) the
preparations for which were of the highest scientific merit ;
but the law itself, I venture to believe, first became a living
fact in science by the publication of my observations made
from 1798 to 1804 in the south of Prance, in Spain, the
Canaries, and the interior of tropical America north and
south of the equator, and in the Atlantic and Pacific Oceans.
The scientific voyages of Le Gentil, Peuille'e, and Lacaille, —
the first attempt to construct an inclination map, by Wilke
in 1768, — and the memorable voyages of circumnavigation of
Bougainville, Cook, and Vancouver, — all deserve honorable
mention for the data they afforded in respect to the previ-
ously much neglected element of the inclination, so impor-
tant for the establishment of the theory of terrestrial
magnetism. The instruments employed were, indeed, of very
unequal merit, and the determinations (which, although
widely distributed over the earth's surface, were generally

64 TEKRESTRIAL MAGNETISM.

limited to the sea and to its immediate vicinity) were far
from being contemporaneous. Towards the end of the 18th
century, the observations of the declination made with
better instruments, at fixed stations, by Cassini, Gilpin, and
Beaufoy (1781 to 1790), showed more decidedly a periodical
influence of hours and seasons, and gave a more animated
and general impulse to magnetic research.

In the 19th century, of which little more than the half
has now elapsed, this branch of scientific inquiry has assumed
a peculiar character distinguishing it from all others. This
character consists in an almost simultaneous advance of all
parts of the study of terrestrial magnetism, in which physical
discoveries relating to the elicitation and distribution of mag-
netism, and the first and brilliant projection of a theory of ter-
restrial magnetism based on strict mathematical reasoning, by
Friedrich Gauss, have accompanied the unprecedented exten-
sion of numerical determinations of all the magnetic elements,
— the declination, inclination, and intensity of the force.

The means which have led to this result have been : —
the improvement of instruments and methods of observation ;
scientific naval expeditions, on a scale and in number such
as no previous century had witnessed, carefully equipped at
the costs of the governments which sent them forth, and
favoured by a happy choice of commanders and observers ;
land journeys, penetrating far into the interior of continents ;
and lastly, the establishment of a considerable number of
fixed observatories, extending partially over both hemi-
spheres, in corresponding north and south latitudes, and in
almost antipodal longitudes.

These magnetical and, at the same time, meteorological
observatories form a kind of net-work over the earth's

TERRESTRIAL MAGNETISM. 65

surface, and by the judicious combination of the observations
made at them, important and unexpected results have been
arrived at. The subjection of the manifestations of the
magnetic force, as displayed on the earth's surface, to
determinate laws, — which subjection, in relation to all
forces, is the proximate, but not the ultimate object of all
scientific inquiry, — has been already satisfactorily esta-
blished and investigated in several particular phases of the
phenomena. In the path of physical experimentation, on
the other hand, the discoveries which have been made of the
relations of terrestrial magnetism to electricity in motion, to
radiant heat, and to light ; the recent generalisation of the
phaenomena of diamagnetism, and the discovery of the
specific property of the oxygen of the atmosphere to acquire
polarity, — all open to us the cheering prospect of a nearer
approach to the nature of the magnetic force itself.

In order to justify the praise which I have ventured to
bestow on the magnetic labours, taken generally, of the
first half of the present century, I append a brief notice of
the more prominent among them, arranging them sometimes
singly in chronologiccil order, and sometimes, where they
appear to have called each other forth, in groups. (68)

1803 — 1806. Krusenstern's Voyage of Circumnaviga-
tion, 1812. The magnetical and astronomical portion of the
work is by Homer. (Bd. iii. S. 317.)

1804. Examination of the law of increase of the intensity
of the earth's magnetic force north and south of the mag-
netic equator, founded on observations made from 1799 to
1804. (Humboldt, Yoyage aux Regions equinoxiales du

Nouveau Continent, T. iii. p. 615 — 623 ; Lametherie,
VOL. IY. F

66 TERRESTRIAL MAGNETISM.

Journal de Physique, T. Ixix. 1804, p. 433, with the first
sketch, of a map of the intensity ; Kosmos, Bd. i. S. 432,
Anm. 29 ; Engl. ed. p. 416, Note 159). Subsequent ob-
servations have shown that the minimum of intensity is not
situated on the magnetic equator ; and that the increase of
force in either hemisphere does not extend to the magnetic
pole.

1805 — 1806. Gay-Lussac and Humboldt's observations
on the magnetic force in the South of France, Italy, Switzer-
land, and Germany (Memoires de la Societe d'Arcueil, T. i.
p. l-r-22.) Compare therewith the observations of
Quetelet, 1830 and 1839, published in the Mem. de
F Academic de Bruxelles, T. xiv., with a map of the hori-
zontal magnetic force between Paris and Naples ; Forbes's
observations in Germany, Flanders, and Italy, in 1832 and
1837 (Transactions of the Eoyal Society of Edinburgh,
Yol. xv. p. 27) ; the very exact observations of Rudberg in
France, Germany, and Sweden, 1832; and the observations
of Bache (Director of the Coast-Survey of the United
States), of inclination and force, at 21 stations in 1837
and 1840.

1806 — 1807. A long series of observations at Berlin
on the horary variations of the declination, and on the
recurrence of " magnetic storms/' (perturbations or dis-
turbances), by Humboldt and Oltmanns, made chiefly at
the solstices and equinoxes for five or six, or sometimes
even nine, successive days and nights, with a Prony's
magnetic telescope reading to 7 or 8 seconds of arc.

1812. Statement by Morichini at Rome that unmagnetised
steel needles became magnetic by contact with light (with
the violet ray). On the long controversy occasioned by this

TERRESTRIAL MAGNETISM. 67

statement, and by Mary SomervihVs ingenious experiments,
down to the wholly negative results of Riess and Moser,
see Sir David Brewster's Treatise on Magnetism, 1837,
p. 48.

1815—1818, and 1823—1826. The two voyages of cir-
cumnavigation of Otho von Kotzebue; the first in the
f Rurik/ the second in the ' Predprijatie/ five years later.

1817 — 1848. The series of great scientific naval expedi-
tions sent by the French government, and which have been
so fruitful in results contributing to the knowledge of terres-
trial magnetism ; beginning with Ereycinet, in the ' Ura-
nie/ 1819 — 1820; followed byDuperrey, in the 'Coquille/
1822—1825; Bougainville, in the 'Thetis/ 1824—1826 ;
Dumont d'Urville, in the 'Astrolabe/ 1826—1829, and in
the Antarctic Regions in the 'Zelee/ 1837—1840; Jules
de Blosseville in India, 1828 (Herbert, Asiatic Researches,
Yol. xviii. p. 4 ; Humboldt, Asie centrale, T. iii. p. 468),
and in Iceland, 1833 (Lottin, Voyage de la Recherche,
1836, p. 376—409); du Petit Thouars (with Tessau), in
the 'Venus/ 1837—1839; Le Vaillant, in the 'Bonite/
1836 — 1837 ; the expedition of the Commission Scientifique
du Nord (Lottin, Bravais, Martins, and Siljestrom) to
Scandinavia, Lapland, the Feroe Islands, and Spitzbergen,
in the corvette 'Recherche/ 1835—1840; Berard, to the
Gulf of Mexico and North America, 1838, and to the
Cape of Good Hope and St. Helena in 1842 and 1846,
(Sabine, in the Phil. Trans, for 1849, Pt. ii. p. 173) : arid
Francis de Castelnau, Voyage dans les parties centrales de
TAmerique du Sud, 1847— 1850.

1818 — 1853. The long series of important expeditions

sent to the Arctic Seas by the British Government., to which

r 2

68 TERRESTRIAL MAGNETISM.

the first impulse was given by the praiseworthy zeal of John
Barrow : Edward Sabine's magnetical and astronomical
observations in John Boss's Voyage to Davis Straits, Baffin's
Bay, and Lancaster Sound, 1818 ; and in Parry's Voyage (in
the ' Hecla' and ' Griper') through Barrow's Straits to Melville
Island, 1819—1820 ; Franklin, Eichardson, and Back, 1819
— r!822; the same, 1825—1827; and Back alone, 1833—
1835. (In the first of these last-named expeditions, almost the
only food for several weeks was a lichen, Gyrophora pmtu-
lata, Tripe de Roche of the Canadian hunters, chemically exa-
mined by John Stenhouse in the Phil. Trans, for 1849, Pt. ii.
p. 393) ; Parry's second expedition with Lyon, in the ' Fury'
and 'Hecla,' 1821 — 1823; and third voyage with James
Clark Ross, in 1824 — 1825 ; Parry's fourth voyage (an at-
tempt to reach the North Pole over the ice to the north of
Spitzbcrgen) with the same, and Lieutenants Foster and
Crozier, 1827, when they reached lat. 82° 45' : John Ross,
with his distinguished nephew James Clark Ross (the ex-
penses of this voyage, which proved so perilous from its
long duration, 1829 — 1833, were defrayed by a private
individual, Felix Booth) ; Dease and Simpson (of the
Hudson's Bay Company) 1838—1839; and recently, in the
search for Sir John Franklin, the voyages of Austin,
Ommaney, and Penny, 1850 — 1851. Of these, Penny
advanced farthest to the north, — to lat. 77° 6' in Victoria
or Queen's Channel, which opens from Wellington Strait.

1819 — 1821. Bellinghausen's voyage in the Antarctic
Seas.

1819. The publication of the great work of Hansteen,
Magnetismus der Erde, which had, however, been finished
as early as 1813. This excellent work has exercised

TERRESTRIAL MAGNETISM. 69

an unmistakable influence in the animation and better
direction of researches in terrestrial magnetism. It was
followed by Hansteen's general maps of lines of equal
inclination and equal force for a considerable portion of the
earth's surface.

1819. Observations of Admiral Roussin and Givry on the
coast of Brazil, between the mouths of the Amazon and the
Eiver Plate.

1819 — 1820. Oersted's great discovery of the fact that a
conductor traversed by an electric current, forming a com-
plete and continuous circuit, exercises, so long as the current
continues, a definite influence on the direction of the mag-
netic needle dependent on their relative position. The
earliest extension of this discovery, as well as of those of the
separation of the metals from the alkalies, and of double
polarisation, the most brilliant of the discoveries of
the age,(69) was Arago's observation that a connecting
wire through which an electric current flows, though made
of copper or platinum, attracts and retains iron filings like a
magnet ; and also, that needles placed inside a galvanic coil re-
ceive opposite poles according as the turns of the coil are given
an opposite direction (Annales de Chimie et de Physique,
T. xv. p. 93). The discovery of these phaenomena, which
were traced under a variety of circumstances, was followed
by Ampere's ingenious theoretical combinations on the
reciprocal electro-magnetic actions of the molecules of pon-
derable bodies. These combinations were supported by
much new and ingeniously-devised apparatus, and led to a
recognition of laws in many phsenomena of magnetism which
had previously appeared contradictory.

1820 — 1824. Wrangel and Anjou's journeys to the North

70 TERRESTRIAL MAGNETISM.

Coast of Siberia and on the Icy Sea. (Important phseno-
mena of the Aurora Borealis : Th. ii. S. 259.)

1820. Scoresby, Account of the Arctic Eegions :
(experiments on the magnetic force : Yol. ii. p. 537 —
554.)

1821. Seebeck's discovery of thermo-magnetism and
thermo-electricity. The contact of two metals (first tried
with bismuth and copper), or differences of temperature at
the two points of contact of a metallic ring, are recognised
as sources of excitement of magneto-electric currents.

1821—1823. Weddell's voyage in the Southern Polar
Sea, to latitude 74° 15' S.

1822 — 1823. Sabine's two important expeditions for the
exact determination of the magnetic inclination and variations
of the magnetic force, and of the length of the pendulum
in different latitudes (west coast of Africa to the Equator,
Brazil, Havanna, Greenland to latitude 74° 23', Norway
and Spitzbergen to latitude 79° 50'). This comprehensive
work was published in 1824 : (Account of Experiments to
determine the Figure of the Earth, p. 460—509.)

1824. Erikson's magnetic observations on the shores of
the Baltic.'

1825. Arago discovered rotation-magnetism. This
unexpected discovery was a consequence of his noticing,
when at Greenwich, that the time required by a dipping
needle, when set in vibration, to come to rest, was influenced
by neighbouring non-magnetic substances. In Arago's rota-
tion experiments the vibrations of a needle were found to be
affected by water, ice, glass, charcoal, and mercury. (70)

1825 — 1827. Boussingault's magnetic observations in
different parts of South America (Marmato, Quito).

TEKRESTIUAL MAGNETISM. 71

1826 — 1827. Keilhau's observations of the magnetic
force at twenty stations (in Finmarken, Spitzbergen, and
Bear Island) ; and Keilhau and Boeck's observations in
in South Germany and Italy (Schum. Astr. Nachr. No.
146.)

1826 — 1829. Lutke's Voyage of Circumnavigation.
The magnetic part was drawn up with great care in 1834
by Lenz. (See Partie nautique du Voyage, 1836.)

1826—1830. Captain Philip Parker King's observa-
tions on the east and west coasts of South America
(Brazil, Monte "Video, Straits of Magellan, Chiloe, and
Yalparaiso).

1827—1839. Quetelet, Etat du Magnetisme terrestre
(Bruxelles) pendant douze annees. Yery exact obser-
vations.

1827. Sabine, Comparison of the relative intensity of
the earth's magnetic force in Paris and in London. An
analogous comparison of the force in Paris and at Christiania
was made by Hansteen, 1825 and 1828 (British Asso-
ciation Eeports, 1837, p. 19 — 23). The numerous results
of French, English, and Scandinavian travellers, on the in-
tensity of the horizontal force, were first brought into nume-
rical connection, so as to afford relative values, by means of
these two comparisons, in which intercompared needles were
vibrated at the three above-named places. The relative
numbers found were : for Paris, 1*348 by me ; for London,
1-372 by Sabine; and for Christiania, T423 by Hansteen.
All are relative to the intensity of the earth's magnetic
force at a point of the " magnetic equator" (or line without
inclination), where it intersects the Peruvian Cordillera,
between Micuipampa and Caxamarca in S. latitude 7° 2',

72, TERRESTRIAL MAGNETISM.

and W. longitude 78° 46', and where the intensity is taken
as=rOOO. It was to this point (Humboldt, Becueil
d'Observ. astr., T. ii. p. 382—385 ; and Voyage aux
Eegions equinox. T. iii. p. 622) that for forty years the
reductions in all tables of the force were referred as a basis
(Gay-Lussac, in the Mem. de la Soc. d'Arcueil, T. i. 1807,
p. 21 ; Hansteen liber den Magnetismus den Erde, 1819,
S. 71 ; Sabine, in the Report of the British Association at
Liverpool, p. 43 — 58.) It has since been justly objected,
that the point so taken as unity does not afford an appro-
priate general standard, since the line of no inclination (71)
does not correspond with the points of weakest intensity
in many meridians, and that no point on the earth's surface
can be taken as a permanent unity, on account of secular
change. (Sabiiie, in the Phil. Trans, for 1846, Part iii.
p. 254 ; and in the Manual of Scientific Inquiry for the use
of the British Navy, 1849, p. 17.)

1828 — 1829. Hansteen and Due's expedition to Siberia :
magnetic observations in European Eussia and Eastern
Siberia, as far as Irkutzk.

1828 — 1830. Adolph Erman's journey through Northern
Asia, and voyages in the Pacific and Atlantic Oceans in
the Eussian frigate Krotkoi. The identity of the instru-
ments used, the employment of the same or similar methods
throughout, the exactness of the astronomical determinations
of geographical position, and the whole of the observations
being made by the same thoroughly informed and practised
observer, returning to the same point after having gone
round the globe, are all circumstances which combine in
assigning a high value to this enterprise, executed at
private expense. (See the General Map of the Declination,

TERKESTRIAL MAGNETISM. 73

founded on Erman's observations, in the Report of the
Committee of Physics of the Royal Society on the occasion
of the Antarctic Expedition, 1840, Plate iii.)

1828 — 1829. Humboldt' s continuation of the observa-
tions at the solstices and equinoxes with a Gambey's needle
on horary variation and extraordinary perturbations, begun
in 1800 and 1807 in a magnetic house built expressly
for the purpose at Berlin. Corresponding determinations
made at Petersburg, NikolajefF, and in the mines at
Ereiberg, by Professor Reich, at 216 feet below the surface.
Dove and Biess continued these observations until Nov.
1830, including both declination and intensity of the hori-
zontal force. (Poggend. Annalen, Bd. xv. S. 318 — 336 ;
Bd. xix. S. 375—391, with 16 Tables; Bd. xx. S. 545—
555.)

1829—1834. The botanist David Douglas, who was
killed at Owyhee by falling into a pit into which a wild bull
had previously fallen, made a fine series of magnetic obser-
vations on the north-west coast of America and in the
Sandwich Islands, including one station on the edge of the
crater of Kiraueah (Sabine, at the British Association
Meeting at Liverpool, p. 27 — 32).

1829. Kupffer, Voyage au Mont Elbrouz dans le Cau-
case (p. 68 and 1]5).

1829. Humboldt, observations on terrestrial magnetism,
together with astronomical determinations of geographical
position, on a journey by command of the Emperor Nicholas
in Northern Asia, from the long, of 11° 3' to that of
80° 12' east of Paris, near Lake Dzaisan, and from the lat.
of 15° 43' (in the Island of Birutschicassa in the Caspian) to

74 TERKESTPJAL MAGNETISM.

58° 52' in the northern part of the Oural Mountains at
Werchoturie. (Asie centrale, T. iii. p. 440—478.)

1829. The Imperial Academy of Sciences at St. Peters-
burg consented to my proposition for the establishment of
magnetical and meteorological stations in the most varied
climatic zones of the Eussian dominions in Europe and
Asia, and for the erection of a physical central observatory
in the capital of that empire, under the active and able
direction of Professor Kupffer. (Compare Kosmos, Bd. i.
S. 436—439, Anm. 36; Engl. p. 419—421, Note 166;
Kupffer, Rapport adresse a FAcad. de St.-Petersbourg relatif
& I'Observatoire physique central fonde aupres du Corps
des Mines, in Schum. Astr. Nachr. No. 726 ; and in the
Annales magnetiques, p. xi.) By the unfailing support
given by Count Cancrine, the Minister of Finance, to e^ery
great scientific undertaking, it was found possible to com-
mence a portion of the corresponding observations (72)
from the Crimea to the White Sea, and from the Gulf of
Finland to the shores of the Pacific in Eussian America, as
early as 1832. A permanent magnetic station was esta-
blished in Pekin in the old convent which had been periodi-
cally inhabited by monks of the Greek Church since the
time of Peter the Great. A highly informed and scientific
observer, the astronomer Fuss, who had taken the chief part
in the measurements for determining the difference of level
between the Caspian and the Black Sea, was selected to make
the first magnetic arrangements in China. Subsequently,
Kupffer visited the magnetical and meteorological stations
as far eastward as Nertschinsk in long. 117° 16' east from
Paris, in order to compare the instruments established at

TERRESTRIAL MAGNETISM. 75

them with the proper standards. The (no doubt) excellent
observations made by Fedoroff, in Siberia, are still unpub-
lished.

1830 — 1845. Colonel Graham (of the topographical
engineers of the United States) : observations on the
magnetic force on the southern boundary of Canada. (Sabine,
in the Phil. Trans, for 1846, Pt. iii. p. 242).

1830. Fuss : magnetic, astronomic, and hypsometric obser-
vations (Report of the Seventh Meeting of the Brit. Assoc*
1837, p. 497 — 499) on a journey from Lake Baikal
through Ergi Oude, Durma, and the. region of Gobi (which
is only about 2500 feet high), to Pekin, in order to
found there the magnetical and meteorological observatory
in which Kovanko has observed for ten years. (Humboldt,
Asie centrale, T. i. p. 8 ; T. ii. p. 141 ; T. iii. pp. 468 and
477.)

1831 — 1836. Captain Fitz Eoy, in a voyage of circum-
navigation in the ' Beagle/ as well as in the survey of the
coasts of the southern extremity of America ; observations
with a Garabey's Inclinatorium, and horizontal needles
received from Hansteen.

1831. Dunlop, Director of the Astronomical Observa-
tory of Paramatta : observations on a Voyage to Australia,
(Phil. Trans, for 1840, Pt. i. p. 133—140).

1831. Faraday's Induction-currents, the theory of which
has been extended by Nobili and Antinori ; great discovery
of the production of light by magnets.

1833 and 1839 are the two important epochs of the first
promulgation of the theoretical views of Gauss: — 1. Intensitas
vis magneticse terrestris ad mensuram absolutam revocata,
1833 (p. 3, "elementum tertium, intensitas, usque ad

76 TERRESTRIAL MAGNETISM.

tempora recentiora penitus neglectum mansit") ; 2. The
immortal work, Allgemeine Theorie des Erdmagnetismus,
in the Resultate aus den Beobachtungen des magnetischen
Vereins im Jahr 1838, herausgegeben von Gauss und Weber,
1839, S. 1—57, (General Theory of Terrestrial Magnetism,
published in English in Vol. ii. of the Scientific Memoirs,
p. 184—251; and Supp. Yol. ii. p. 313—317).

] 833. Barlow's investigations on the attraction of ships'
iron, and the means of determining its deflecting influence
on the compass ; examination of electro-magnetic currents
in " Terellas." General Maps of the Declination. (Compare
Barlow's " Essay on Magnetic Attraction," 1833, p. 89,
with Poisson, " sur les deviations de la boussole produites
par le fer des vaiseaux," in the Mem. de 1'Institut, T. xvi. p.
481—555; Airy, in the Phil. Trans, for 1839, Pt. i. p. 167,
and for 1843, Pt. ii. p. 146 ; Sabine, in the account of Sir
James Ross's observations, in the Phil. Trans, for 1849
Pt.ii. p. 177— 195.)

1833. Moser, in PoggendorfFs Annalen, Bd. ii. S. 53 —
64, on a method of ascertaining the position and force of
the variable magnetic poles.

1833. Christie on the Arctic observations of Captain
Back, Phil. Trans, for 1836, Pt. ii. p. 377. See also his
earlier and important Memoir in the Phil. Trans, for 1825,
Pt. i. p. 26.

1834. Parrot's Eeise nach dem Ararat (Journey to Mount
Ararat), Magnetismus, Bd. ii. S. 33 — 64.

1836. Major Estcourt in Colonel Chesney's Euphrates
expedition. A part of the Force observations was lost in
the steamer ' Tigris,' a circumstance which is the more to
be regretted because there is so entire an absence of exact

TERRESTRIAL MAGNETISM. 77

observations in this part of Western Asia, and generally
south of the Caspian.

1836. Lettre de M. de Humboldt k S. A. R. le Due de
Sussex, President de la Soc. Roy. de Londres, sur les
moyens propres a perfectionner le connaissance du magne-
tisme terrestre par Fetablissement de stations magnetiques
et d' observations correspondantes (Avril, 1836). On the
happy results of this application, and its influence in
contributing towards the great Antarctic Expedition of Sir
James Boss, see Kosmos, Bd. i. S. 438 (Engl. p. 421) ;
and Sir James Ross's Voyage to the Southern and Antarctic
Regions, 1847, Vol. i. p. xii.

1837. Sabine on the variations of the intensity of the
magnetic force of the earth : British Association Reports,
Liverpool Meeting, p. 1 — 85. This is the most complete
work which has been published on this branch of the subject.

1837 — 1838. Establishment of a magnetic observatory
at Dublin by Professor Humphry Lloyd. On the observa-
tions made there from 1840 to 1846, see Trans, of the
R. LA. Vol. xxii. Pt. i. p. 74—96.

1837. Sir David Brewster : A Treatise on Magnetism,
p. 185—263.

1837 — 1842. Sir Edward Belcher's observations in a
voyage to Singapore, the Chinese Seas, and the West Coast
of America : Sabine, in the Phil. Trans, for 1843, Pt. ii. p.
113, 140, 142. The observations of the inclination viewed
in connection with mine in 1803 indicate a very unequal
progression in the secular change at different places. I
found, for example, the inclinations at Acapulco, Guayaquil,
and Callao + 38° 48', + 1 0° 42', and - 9° 54' ; Sir Edward
Beicher, + 37° 67', + 9° 1', and -9° 54'. May the frequent

78 TERRESTRIAL MAGNETISM.

earthquakes on the Peruvian coast exert a local influence on
the phenomena of the earth's magnetic force ?

1838—1842. Wilkes's Narrative of 'the United States
Exploring Expedition (Vol. i. p. xxi.)

1838. Lieutenant Sulivan's observations in a voyage from
Palmouth to the Falkland Islands : Sabine, in the Phil.
Trans, for 1840, Pt. i. p. 129, 140, and 143.

1838 and 1839. Establishment of the magnetic obser-
vatories in the two hemispheres at the cost of the British
Government, and under the able direction of Colonel Sabine.
The instruments were sent out in 1839; the observations
began at Toronto in Canada, at Hobarton in Yan
Diemen Island, and at St. Helena, in 1840, and at the
Cape of Good Hope in 1841. (Sir John Herschel in the
Quarterly Review, Yol. Ixvi. 1840, p. 297; Becquerel,
Traite d'Electricite et de Magnetisme, T. vi. p. 173). By
a laborious and profound investigation and treatment of the
rich treasure of observations obtained from these stations,
embracing all the elements or variations of the earth's
magnetic activity, Colonel Sabine, as Superintendent of the
Colonial Observatories, has discovered laws previously un-
known, and opened new views to science. The results of
his investigations have been published by him in a long
series of memoirs, entitled Contributions to Terrestrial
Magnetism, Nos. I. to IX. (and other detached memoirs)
in the Phil. Trans, as well as in separate volumes, and
constitute an essential part of the foundation of this branch
of cosmical knowledge. I will name only some of the most
remarkable among them. 1. Observations on days of
unusual magnetic disturbance, Yol. i. 1840 to 1844 inclu-
sive; and Phil. Trans, for 1851, Pt. i. p. 123—139.

TERRESTRIAL MAGNETISM, 79

2. Observations made at the magnetical observatory at
Toronto (N. lat. 43° 39' ; W. long. 79° 21'-5), Yol. i. 1840,
1841, and 1842; and Yol. ii. 1843, 1844, and 1845-

3. The very different march of the magnetic declination in
the two half-years at St. Helena $. lat. 15° 55', W. long.
5° 41') : Phil. Trans, for 1847, Pt. i. p. 54. 4. Observa-
tions made at the magnetical and meteorological observatory
at the Cape of Good Hope, 1841 — 1846: Yol. i. Magnetism.

5. Observations made at the magnetical and meteorological
observatory at Hobarton (8. lat. 42° 52', E. long. 147° 27''5)
in Yan Diemen Island, and on the Antarctic Expedition,
Yols. i., ii., and hi., 1841 — 1852 (on the separation of the
eastern and western disturbances, see Yol. ii. p. ix. — xxxvi.)

6. Magnetic phenomena in the Antarctic circle, and in Ker-
guelen and Yan Diemen Islands : Phil. Trans, for 1843,
Pt. ii. p. 145 — 231. 7. On the Isoclinal and Isodynamic
Lines in the Atlantic Ocean in 1837 : Phil. Trans. 1840,
Pt. i. p. 129 — 155. 8. Declination Map of the Atlantic
Ocean, representing the lines of magnetic declination
between 60° N. and 60° S. latitude for 1840 : Phil. Trans.
1849, Pt. ii. p. 173—233. 9. On the means adopted at
the British Colonial observatories for determining the abso-
lute values, secular change, and annual variation of the
magnetic force : Phil. Trans. 1850, Pt. i. p. 201—219.
(Coincidence shown of the epoch of the Earth's greatest proxi-
mity to the Sun, with the greatest intensity of the terrestrial
magnetic force in both hemispheres, and with the greatest
amount of inclination, p. 216.) 10. On the isoclinal and
isodynamic lines in the northern parts of the North Ame-
rican continent, and on the geographical position of the
point of maximum force, deduced from the observations of

80 TERRESTRIAL MAGNETISM.

tions of Captain Lefroy : Phil. Trans. 1846, Pt. in.
p. 237—336. 11. On the periodic laws of the disturb-
ances of the declination (magnetic storms) at Toronto in
Canada, and at Hobarton in Van Diemen Island, and on
the accordance of the approximately decennial period of the
magnetic variations depending on the sun, with the also
approximately decennial period, discovered by Schwabe of
Dessau, in the phenomena of the solar spots : Phil. Trans.
1852, Pt. i. p. 121—124.

1839. Isoclinal and isodynamic lines in the British
Islands, from observations of Humphry Lloyd, John
Phillips, Robert Were Eox, James Eoss, and Edward
Sabine. In 1833, the British Association, at Cambridge,
pointed out the importance of systematic observations of
the inclination and force being made in different parts of
the kingdom; and already, in the summer of 1834, their
wish had begun to be fulfilled by Professor Lloyd and
Colonel Sabine; in 1835 and 1836 the work was extended
to Wales and Scotland, and in 1838 complete isoclinal and
isodynamic maps of the British Islands were presented to
the British Association, and published in the Report of the
Meeting at Newcastle, p. 49 — 196.

1838—1843. The important voyages of Sir James Clark
R/oss towards the South . Pole, equally admirable for the
knowledge gained thereby of the existence of the much-
doubted Antarctic lands, and for the new light thrown on
the magnetic condition of very -large and important portions
of the earth's surface. The numerical values of the three
elements of terrestrial magnetism were determined over
almost two-thirds of the area of the higher latitudes of the
Southern hemisphere. Sir James Ross's magnetic obser-

TERRESTRIAL MAGNETISM. 81

vations have been co-ordinated and published by Colonel
Sabine in the Phil. Trans, for 1843, Art. x., and for 1844,
Art. vii. ; and part yet remains to be published.

1839 — 1851. Kreil's observations on the variations of
the three magnetic elements, continued for more than
twelve years at the Imperial Astronomical Observatory at
Prague, and published in annual volumes, with valuable
co-ordinations and discussions.

1840. Hourly magnetic observations with a Gambey's
declination-needle, made during a ten years' sojourn in
Chili by Claudio Gay. (See his Historia fisica y politica
de Chile, 1847.)

1840 — 1851. Lament, Director of the Astronomical
Observatory at Munich : Results of his magnetic obser-
vations compared with those of Gottingen, which go back
as far as 1835 : important law of a decennial period in the
diurnal variation of the declination. (Lamont, in Poggend.
Ann. der Phys. 1851, Bd. Ixxxiv. S. 572—582; and Eels-
huber, 1852, Bd. Ixxxv. S. 179— 184.) The already noticed
conjecture of a connection between the periodical magnetic
variations which follow laws depending upon solar hours
(magnetic storms and diurnal variation), and the periodical
" frequency of the solar spots," was first made by Col. Sabine,
in the Phil. Trans, for 1852 ; and four or five months later,
but without his having known of Colonel Sabine' s published
paper, by the learned Director of the Astronomical Obser-
vatory at Berne, Rudolph Wolf, in the Schriften der
schweizerischen Naturforscher.(73) Lamont's Handbuch
des Erdmagnetismus (1848) contains an account of the
most recent German apparatus and methods of computing
the results of observation.

VOL. IV. Q

82 TERRESTRIAL MAGNETISM.

1840—1845. Bache, Director of the Coast Survey of
the United States : Observations made at the Ma<metical

O

and Meteorological Observatory at Girard's College (Phila-
delphia), published 1847.

1840—1842. Lieutenant Gilliss (U.S.): Magnetical and
meteorological observations made at Washington, pub-
lished 1847 (p. 2—319; magnetic storms, p. 336).

1841—1843. Sir Robert Schomburgk : Declination ob-
servations in the forest region of Guyana between Mount
Eoraima and the village of Pirara, between the parallels of
4° 57' and 3° 39'. (Sabine, in the Phil. Trans, for 1849,
Partii. p. 217.)

1841 — 1845. Magnetical and meteorological observa-
tions made at Madras.

1843^1844. Magnetic observations at Sir Thomas
Brisbane's observatory at Makerstoun (Roxburghshire,
Scotland), in latitude 55° 34'. (See Transact, of the Royal
Society of Edinburgh, Yol. xvii. Part ii. p. 188; and
Yol. xviii. p. 46.)

1843—1849. Kreil on the influence of the Alps on the
manifestations of the terrestrial magnetic force. (Com-
pare Schum. Astr. Nachr. No. 602.)

1844—3845. Expedition of the ' Pagoda' to high Ant-
arctic latitudes (64°-67°), E. longitude 2° to 115°, em-
bracing all the three elements of terrestrial magnetism, under
the command of Lieut. Moore, R.N., who had previously
been emploped in the Antarctic Expedition ; the observa-
tions being made jointly by him and Lieut. Clerk, of the
Royal Artillery, who had previously directed the magnetic
observatory at the Cape of Good Hope ; — a worthy comple-
tion of the Antarctic labours of Sir James Clark Ross.
(Sabine, in Phil. Trans. 1846).

TERRESTRIAL MAGNETISM. 83

1845. Proceedings of the Magnetical and Meteorological
Conference held at Cambridge.

1845. Observations made at the Magnetical and Me-
teorological Observatory at Bombay, under the superin-
tendence of A. B. Orlebar. The observatory was built in
1841, on the little island of Colaba.

1845 — 1850. Six volumes of Eesults of the Magnetical
and Meteorological Observations made at the Eoyal Ob-
servatory at Greenwich. The magnetic observatory was
built in 1838.

1845. Simonoff, Prof, at Kazan : Eecherches sur F Action
magnetique de la Terre.

1846—1849. Captain Elliot (Madras Engineers) : Mag-
netic Survey of the Eastern Archipelago, sixteen stations,
several months being passed at each, in Borneo, Celebes,
Sumatra, the Nicobar, and the Keeling Islands, and
extending between N. latitude 16° and S. latitude 12°,
longitudes 80° and 125° E. (Phil. Trans, for 1851, Part i.
p. 287 — 331, and p. 1. — clvii.) There are appended maps
of the lines of equal inclination and declination, as well
as of the horizontal and total force. This work, which
also represents the positions of the magnetic equator, or
line of no inclination, and of the line of no declination, is
one of the most remarkable and comprehensive of modem
times.

1845 — 1850. Faraday's brilliant physical discoveries :
1. On axial (paramagnetic) and equatorial (diamagnetic)
direction assumed by different bodies when freely sus-
pended and placed under external magnetic influences. (74)
(Phil. Trans, for 1846, § 2420 ; and Phil. Trans, for
1851, §2718—2796); 2. On the relation of electro-

G 2

84 TERRESTRIAL MAGNETISM.

magnetism to a polarised ray of light, and the rotation of
the polarised ray under the intervention of an altered mole-
cola r condition of the substance which is made the conduct-
ing medium both of the ray and of the magnetic current
(Phil. Trans, for 1846, § 2195 and 2215—2221);
3. On the remarkable property possessed by oxygen gas
(the only one of the gases which is paramagnetic) similar to
that possessed by soft iron only in a much weaker degree,
of assuming polarity under the inducing influence of the
earth, as soft iron does in the presence of a permanent
magnet(75). (phfl. Trans. 1851, § 2297—2967.)

1849. Professor William Thompson, of Glasgow : A
mathematical theory of magnetism, Phil. Trans, for 1851,
Part i. p. 243—285. (On the problem of the distribution
of the magnetic force, compare § 42 and 56, with Poisson,
in the Mem. de 1'Institut, 1811, Part i. p. 1; Part ii.
p. 163.)

1850. Airy on the present state and prospects of the
Science of terrestrial magnetism, a fragment of a very
promising essay.

1852. Kreil, Influence of the Moon on the magnetic
declination at Prague in the years 1839 — 1849. On the
earlier labours of this exact observer, and on his first con-
jectures regarding the lunar influence, see Osservazioni sullj
intensita e sulla direzione della forza magnetica istituite negli
anni 1836— 1838, all' I. E. Osservatorio diMilano, p. 171 ;
and his Mag. und Met. Beobachtungen zu Prag, Bd.' i.
S. 59.

1853. Faraday on lines of magnetic force, and their
definite character.

1853. Sabine: on the lunar diurnal variation of the

TERRESTRIAL MAGNETISM. 85

Declination at Toronto, St. Helena, and Hobarton. (Phil.
Trans. 1853.)

1853—1854. Sabine's new proofs from the observations
at Toronto, Hobarton, St. Helena, and the Cape of Good
Hope (from 1841 to 1851), of the character of the annual
variation superimposed upon the mean diurnal variation of
the Declination, and of the correspondence of its semi-
annual epochs with those of the Sun's passage of the Equator.
(Toronto Observations, Yol. ii. p. xxii. ; and Proceedings of
the Royal Society of London, May 1854.)

The preceding chronological enumeration of the advances
which have been made in our knowledge of terrestrial mag-
netism in the course of the last half-century (during which
I have myself uninterruptedly devoted the warmest interest
to the subject) shows that they have had a two-fold character.
The most considerable portion of these labours is that which
has been devoted to the observation of the manifestations
of the earth's magnetic activity, and to their variations
according to time and place ; the smaller portion has been
given to experiments, or to the calling forth of phenomena
promising to conduct us to a comprehension of the essential
nature of the activity so manifested. These two lines of
research, — on the one hand, that of observation of the
manifestations of terrestrial magnetism (in direction and
force), and on the other, that of physical experimentation
on magnetic force in general, — have acted and reacted on
each other, and have concurrently animated and carried
forward our knowledge of Nature. Observation alone, —
independent of any hypothesis on the causal connection of
the phenomena, or on the as yet immeasurable and

86 TERRESTRIAL MAGNETISM.

unsearchable reciprocal action of molecules hi the interior
of substances, — has led to important numerical laws ; and
the admirable sagacity of experimenting physicists has
succeeded in discovering, in solid and gaseous substances,
previously wholly unsuspected polar properties peculiarly
connected with temperature and with atmospheric pressure.
Important and undoubted as are these discoveries, they
cannot, in the present state of our knowledge, be yet
regarded as affording satisfactory bases of explanation of
the laws derived from the observations of the terrestrial
magnetic phsenomena. The surest means of exhausting the
measurable variations as respects space, and of enlarging
and completing the mathematical theory of terrestrial mag-
netism, so grandly sketched by Gauss, is to prosecute
continuously successive determinations of the three mag-
netic elements at well-chosen points of the globe, each such
determination referring to a definite epoch. The anticipa-
tions which I have myself formed of the great results which
will follow hereafter from the combination of experiment
with mathematical reasoning, have been already touched
upon in the earlier pages of the present volume. (76)

We look in all the physical phsenomena which take place
on our planet, for cosmical connection. The word Planet
of itself leads us to dependence on a central body, and to
connection with a group of heavenly bodies of very various
magnitudes, probably having all the same origin. The
influence of the Sun's position in the heavens on the
manifestation of the Earth's magnetic force was very early
recognised; — most clearly, in the discovery of an horary
variation in the declination of the needle ; more obscurely,
in the conjecture formed by Kepler, a century before, of the

TERRESTRIAL MAGNETISM. 87

axes of all the planets being directed towards the same
quarter. Kepler said expressly, " that the sun may be a
magnetic body, and that the force which moves the planets
may therefore reside in the sun." (77) Mass-attraction
and gravitation appear here under the symbol of magnetic
attraction. Horrebow, (78) who did not confound gravi-
tation with magnetism, was the first who termed the solar
light "a perpetual Aurora in the sun's atmosphere, pro-
duced by magnetic forces." In more modern times (and
the difference is important), the views which have been
formed respecting the modus operandi of the sun's action
have diverged into two distinct classes.

It has been considered either that the sun, without being
itself magnetic, acts on the magnetism of the earth solely
through the medium of its effects in causing variations of
temperature (this has been the view entertained by Canton,
Ampere, Christie, Lloyd, and Airy), — or, as supposed by
Coulomb, that the sun being enveloped in a magnetic atmo-
sphere^9) influences the magnetism of the earth by direct
action. Earaday's striking discovery of the paramagnetic
property of oxygen would indeed remove one great difficulty
opposing the first-named view, i. e. the difficulty of supposing
with Canton that the sea and the solid crust of the earth
have their temperature instantaneously and considerably raised
by the passage of the sun over the meridian of the place ;
but on the other hand, Colonel Sabine's very extensive
combination and sagacious discussion of all the facts of
observation have led to the result, that the periodical varia-
tions in the earth's magnetic activity do not correspond to,
and therefore cannot be assumed to be caused by, the peri-
odic variations of temperature which take place within the

TERRESTRIAL MAGNETISM.

the atmosphere accessible to our observation. Neither
the principal epochs of the diurnal and annual variations of
the declination (the annual have been accurately represented
for the first time by Sabine from a vast body of observa-
tions), nor the periods of the mean intensity of the earth's
force, (80) agree with the periods of maxima or minima in
the temperature either of the atmosphere or of the
earth's outer crust. The turning-points in the most
important magnetic phenomena are the solstices and the
equinoxes.

The epoch at which in loth hemispheres the intensity
of the earth's force is greatest, and the direction of the
dipping-needle is most nearly vertical, is that of the
greatest proximity of the Earth to the Sun,(81) when also
the Earth has the greatest velocity in its movement of trans-
lation in its orbit. But at the period of perihelion
(December, January, and February), and at that of aphelion
(May, June, and July), the circumstances of temperature
in the two hemispheres, northern and southern, are diame-
trically opposed to each other : it follows, therefore, that the
turning-points, or change from decreasing to increasing
magnetic force, and vice-versa, cannot be ascribed to the
sun as the source of heat.

Annual mean values derived from the observations at
Munich and Gottingen have led the active director of the
Royal Bavarian Observatory, Lament, to infer the remark-
able law of a variation-period of 10 J years in the amount of
the mean diurnal variation of the magnetic declination in
different years. (82) In a series lasting from 1841 to 1850,
the monthly means of the variations of the declination
attained their minimum in 1843^ and their maximum in

TERRESTRIAL MAGNETISM. 89

1848-J-, and this in a very regular manner. Without his
being aware of these results, the comparison of the monthly
means of the same years, 1843 — 1848, derived from obser-
vations taken at places separated from each other by almost
an entire diameter of the globe (Toronto in Canada, and
Hobarton in Yan Diemen Island), had led Colonel Sabine
to infer the existence of a similar variation-period in the
frequency and amount of the magnetic disturbances or
storms. He suggested that a purely cosmical cause might
be found in the perturbations observed to take place in the
sun's atmosphere with a frequency varying according to the
same approximately decennial period, f3) Schwabe, the
most diligent observer of the solar spots among astronomers
now living, had found (as I have related in a preceding
volume), (84) in a long series of years, from 1826 to 1850,
a periodical variation in the frequency of the spots, the
maxima falling in the years 1828, 1837, and 1848; and
the minima in the years 1833 and 1843. "I have not/'
he adds, " had the opportunity of examining a continuous
series of older observations, but I willingly subscribe to the
opinion that this period may itself be a variable one." An
analogy to such " periods within periods" as are here
supposed, is presented to us in the luminous processes of
other self-luminous celestial bodies or suns, as in the very
complicated periodicity of the variations in the light of
/3 Lyra3 and Mira Ceti, investigated by Goodricke and
Argelander.(85)

If we consider with Sabine, that the magnetism of the
Sun manifests itself in the increased magnetic force of the
Earth at the time of her perihelion (i. e. when she is nearest
to the Sun), we may be the more surprised that, according

90 TERRESTRIAL MAGNETISM.

to KreiTs extensive investigation of the Moon's magnetic
influence no sensible difference according to the varying
phases of the Moon, or to her different distances from the
Earth, has yet been found. It would seem as if the near
proximity of the Moon, as compared with the Sun, does not
compensate for the smallness of her mass. The principal
result (86) of the examination which has been made into the
magnetic influence of our satellite, which, according to
Melloni, shows only a faint trace of thermal influence, is
that the magnetic declination observed at the surface of our
globe undergoes, in the course of a lunar day, a regular
variation, attaining two maxima and two minima. Kreil
remarks very justly, that "if the Moon produces at the
surface of the Earth no sensible increase of temperature"
(t. e. no increase cognisable by our ordinary thermometers),
" she cannot produce any variation in the Earth's magnetic
force by thermal agency ; and if, therefore, we do, neverthe-
less, find such a variation, we must conclude that it is
produced by some mode of action other than thermal." In
all actions which do not present themselves at once as the
unmistakable results of a single cause, we can only, as here
in the case of the Moon, recognise their independent exist-
tence after carefully excluding, or eliminating, many extra-
neous disturbing elements.

But although at the present time the magnetic variations
of largest amount, and of most decided character, are not, it
must be admitted, satisfactorily explained by the maxima
and minima of the variations of temperature, it is yet scarcely
possible to doubt that when, at some future but- not dis-
tant period, there shall be a more complete and deeper
insight into the whole process of magnetic action, the great

TERRESTRIAL MAGNETISM. 91

discovery of the polar properties of the oxygen of the atmo-
sphere which surrounds our globe will present an element
in the explanation which shall then be given of the genesis
of that process. It is hardly conceivable that in the
harmonious concurrent action of all natural forces this
property of oxygen, and its modification by changes of
temperature, should have no share in calling forth magnetic
phenomena. If, in accordance with the conjecture expressed
by Newton, the substances belonging to the same planetary
system are, in all probability, for the most part the same,(87)
we may also be led, by inductive reasoning, to surmise that
the endowment of gravitating matter with electro-magnetic
activity is not confined to our own globe. To assume the
contrary, would be to impose arbitrarily dogmatic limits on
cosmical views. Coulomb's hypothesis respecting the influ-
ence of the magnetic sun on the magnetic earth is certainly
not opposed to any analogy furnished by the knowledge we
have as yet acquired.

If we now pass to the purely objective representation of
the magnetic phenomena presented to us by the Earth, at
the different parts of its surface, and in its different positions
relatively to the Sun, we must distinguish, in the numerical
results of observation, the variations which are included
within short periods, from those which extend over very long
periods. All are intermingled and intersecting, like circles
of undulation in fluids to which a movement is given, some-
times reinforcing, and sometimes partially or wholly com-
pensating and destroying each other.

In the geographical distribution of the phsenomena, there
present themselves more particularly to our notice : —

92 TERRESTRIAL MAGNETISM.

Two magnetic poles, one in each hemisphere, at un-
equal distances from the earth's poles of rotation : these
are points on the earth's surface at which the magnetic
inclination is 90°, and at which, therefore, the horizontal
force vanishes.

The magnetic equator: i. e. the curve or line encompass-
ing the earth, on which the inclination of the needle is 0°.

Lines of equal declination, and on which the decima-
tion of the needle is 0° (isogonic lines, and lines of no
declination) .

Lines of equal inclination (isoclinal lines) .

Eour points of greatest intensity of the earth's
magnetic force ; two, of unequal strength, in each hemi-
sphere.

Lines of equal terrestrial magnetic force (isodynamic
lines) .

The undulating line which connects those points at
which the force is weakest in each meridian, and which
has, on that account, been termed a " dynamic equator,"
or " equator of force." (88) Tt does not coincide either
with the geographical or the magnetical equator.

The boundaries of the zone (generally of very weak
magnetic force), in which the horary (or diurnal) varia-
tions of the needle at certain hours of the day conform,
during one part of the year to the diurnal variation of
the phsenomena in the northern, and during the other
part of the year to those of the southern, magnetic
hemisphere ;(89) taking part, therefore, alternately in the
variations of both hemispheres.

In this enumeration I have reserved the word "pole"

TERRESTRIAL MAGNETISM. 93

exclusively for the two points on the earth's surface at
which the horizontal force vanishes ; because (as has been
already remarked) these points, in which the maxima of
force are not situated, have occasionally, in modern times,
been confounded with the four points of maximum intensity
of the earth's force. (9°) Gauss has also shown that it is
objectionable to give the name of "magnetic axis of the
earth" to the cord connecting the points above-named, at
which the inclination of the needle at the surface of the
globe is 90°. (91) The various points and lines which
have been thus noticed are so connected with each other,
that we are enabled to describe the complicated phenomena
of terrestrial magnetism under three heads only, or as
belonging to three " elements" or manifestations of one and
the same activity, or force ; termed respectively, — magnetic
"force," "inclination," and "declination."

Intensity of the Magnetic Force.

The recognition and examination of this, the most
important element of terrestrial magnetism, by the direct
measurement of the strength of the total force, did not
follow, until after considerable delay, the recognition and
examination of the direction of the same force measured in
the horizontal and vertical planes (declination and inclina-
tion). Experiments in which a magnetic needle is made to
vibrate for the purpose of inferring from the duration of the
vibrations the intensity of the magnetic force, were first
brought into use towards the close of the last century:
during the present century they have been constantly and
earnestly prosecuted. Graham, in 1723, measured the

94 TERRESTRIAL MAGNETISM.

vibrations of his dipping-needle, with the view of examining
whether they were constant, (92) and of ascertaining the
ratio of the directive force to the force of gravity. The
first attempt to determine, by the number of vibrations
performed in equal times, the comparative force of magne-
tism at points of the earth distant from each other, was
made by Mallet, in 1769. "With very imperfect instru-
ments, he found no difference in the number of vibrations
performed in the same interval of time at St. Petersburg
in lat. 59° 56', and at Ponoi, in lat. 67° 4',p3) whence
arose the erroneous opinion, which continued to the time of
Cavendish, that the intensity of the earth's magnetic force
is the same in all latitudes. Borda, indeed, as he has often
related to me, on theoretical grounds never shared in this
error, nor did Le Monnier ; but Borda himself, in his expe-
dition to the Canaries in 1776, was prevented by the
imperfection of his dipping-needle (friction on its pivots)
from discovering any differences of magnetic force at places
distributed over an interval of 35° of latitude, — Paris,
Toulon, Santa Cruz in Teneriffe, and Goree (Voyage de la
Perouse, Yol. i. p. 162). By the aid of improved instru-
ments, these differences were found for the first time in the
ill-fated expedition of La Perouse, in the years 1785 and
1787, by Lamanon, and were communicated by him from
Macao to the Secretary of the Paris Academy ; but, as I have
already remarked (p. 63), they remained till a much later
period, unnoticed, and buried in the Archives of the
Academy.

The first published observations of the variation of the
force (which were also begun at Borda' s request) were my
own, made in my travels to the equinoctial regions of

TERRESTRIAL MAGNETISM. 95

America in the years 1798 — 1804. Still earlier results on
the earth's magnetic force had been obtained by my friend
De Rossel in 1791 and 1794 in the Indian Seas, but they
were not printed until four years after my return from Mexico.
In 1829 I had the advantage of being enabled to add to my
determinations of the magnetic force and inclination, so as
to make them extend over 188° of longitude (or more than
half the circumference of the globe), viz. from the shores of
the Pacific Ocean eastward to the Siberian frontier of the
Chinese territories, two-thirds of the interval being across
the interior of continents. The differences of latitude were
from 60° N. to 12° S., or 72°.

If, in examining the distribution of the magnetic force over
the surface of the earth, we consider carefully the direction of
the successive isodynamic lines (or curves of equal magnetic
force) enclosing each other, and proceed from the outer or
weakest force to the inner curves of successively stronger
force, we find in each hemisphere, at very unequal distances
both from the poles of the earth and the magnetic poles,
two points, or, as they have been called, " foci," of maximum
force, one stronger and the other weaker. Of these four
points on the earth's surface, we find, taking the northern
hemisphere first, (94) the strongest (the American focus) in
lat. 52° 19', and long. 910^58 W. ; and the weaker (often
termed the Siberian focus) in lat. 70° N., long. 120° E.,
or perhaps a few degrees less easterly. Erman, in 1829, in
travelling from Parchinsk to lakutzk, found the curve of
greatest intensity (1*742) near Beresowski Ostrow, in long,
117° 53' E., lat. 59° 44' N.) ; Erman, Magnet. Beob.
S. 172 and 540 ; Sabine, in the Phil. Trans, for 1850
(Pt. i. p. 218). Of these two determinations, that of the

yb TERRESTRIAL MAGNETISM.

American focus is the more secure, particularly in respect to
latitude : " the longitude," it is said, " is probably rather
too westerly ;" it places the oval which encloses the stronger
of the two northern foci in the meridian of the western end
of Lake Superior. We are indebted for this determination
to the important land expedition in 1843 of Captain
Lefroy, of the British Eoyal Artillery, formerly Director of
the St. Helena, and since of the Toronto, Magnetic Obser-
vatory. " The junction of the two loops of the lemniscate
appears to be situated north-east of Behring's Straits, some-
what nearer to the Asiatic than to the American focus."

When, in 1802, on the Peruvian chain of the Andes
(between Micuipauipa and Caxamarca, in lat. 7° 2 S. and
long. 81° 8' W. from Paris), I intersected the magnetic
equator or line on which the inclination is 0, and when I
found the magnetic force increase as I proceeded either to
the north or to the south from this remarkable spot, — there
being at that time, and long afterwards, an entire absence
of all other points of comparison, — I was led by an erroneous
generalisation of these, the only data which observation
then presented, to believe that the earth's magnetic force
would be found to increase continuously from the magnetic
equator to each of the magnetic poles, or points of 90°
inclination, which I thought were probably also the points
of greatest terrestrial magnetic force. When we come for
the first time on the trace of a great natural law, the views
first taken almost always require some subsequent rectifica-
tion. Sabine(95) has shown, by his own observations (made
through an extensive range of latitude, in the years 1818 to
1822), and by a careful combination of these with those of
many other observers (experiments of vibration with vertical

TE11RESTRIAL MAGNETISM. 97

and horizontal needles having become more and more
general), that the magnetic force and the inclination undergo
very dissimilar modifications, — that the minimum of force
is in many places considerably distant from the line of no
dip ; and even, that in the northern part of Canada and
the Hudson Bay territories, in the meridian of about 92° or
93° W. long., the intensity of the magnetic force, instead
of increasing, decreases from about the lat. of 52° to the
magnetic pole in lat. 70°. At the Canadian focus of
greatest magnetic force in the northern hemisphere, deter-
mined from Lefroy's observations, the inclination of the
needle, in 1845, was only 73° 7', and in both hemispheres
the maxima of magnetic force are found associated with
inclinations much less than 90°.(96)

Excellent and abundant as are the observations of mag-
netic force which we owe to the expeditions of Sir James
Ross, and Captains Moore and Clerk, in the Antarctic Seas,
yet there still remains much doubt respecting the positions
of the stronger and the weaker focus in the southern hemi-
sphere. James Eoss crossed the isodynamic curves of
highest value in several places, and after a careful discussion
of his observations Sabine places the one focus about the
lat. of 64° S.,.long. 137° 30' E. Eoss himself, in the
narrative of his great voyage, (97) supposes the one focus to
be in the vicinity of the Terre Adelie discovered by .D'Ur-
ville, or in about lat. 67° S., long. 140° E. He thought
he was approaching the other focus in 60° S. lat., and 125°
"W. long. ; but was afterwards inclined to place it consider-
ably farther to the south, nearer the magnetic pole, and
therefore in a more easterly meridian. (98)

VOL. IV. H

SS TERRESTRIAL MAGNETISM.

After assigning the positions of the four maxima of
magnetic force, it remains to give the ratio of their
respective degrees of intensity. These are stated either
according to the old relative and arbitrary scale which
has been so often alluded to, i. e. by comparison with the
intensity found by me at the above-mentioned point of the
magnetic equator in lat. 7° 2' S., and long. 81° 8' W. from
Paris, — or in absolute measure, as first proposed by Poisson
and Gauss ("). In the relative scale in which the force
observed by me on the magnetic equator is taken as
= 1*000, the magnetic force in Paris and in London, (em-
ploying the determination of the ratio of the force in Paris
to that in London made in 1827, and mentioned in
p. 71 of the present volume), is 1*348 in Paris, and
1*372 in London. These numbers converted into the abso-
lute scale would be about 10*20 and 10*38 ; and the force
observed by me in Peru, and taken as =1'000 in the arbi-
trary scale, would be, according to Sabine, 7*57 ; this is
higher, therefore, than the force at St. Helena, which is 6*4
in the same absolute scale. All these numbers have still to
undergo further alterations on account of the different years
in which the comparisons were made : in both scales, the rela-
tive or arbitrary, and the absolute (the latter being the pre-
ferable scale), the above data are to be regarded simply as pro-
visional; but even with their present imperfect degree of
exactness they cast a clear light on the distribution of the
terrestrial magnetic force, — an element concerning which,
half a century ago, the most entire ignorance prevailed.
They have also the very great cosmical value of presenting
historic points of commencement from whence to date those

TERRESTRIAL MA.GNETISM 99

changes which future centuries shall manifest, — changes
depending, it may be, on the influence produced by the mag-
netic force of the sun on the magnetism of the earth.

In the northern hemisphere the determination of the force
at the strongest or Canadian focus (lat. 52° 19', long.
92° W.), by Lefroy, is the most satisfactory. It is expressed
in the relative scale by 1*878, the force in London being
1-372; and in the absolute scale by 14*21(100). Even so
far south as New York, in lat. 40° 42', Sabine had found
the magnetic force not much less than at the maximum,
viz. 1-803. At the weaker northern focus, the Siberian
one (lat. ? 70°, long. 120° E.), the force was found, in the
relative scale, by Erman, 1'74, and by Hansteen, 1'76; or,
expressed in absolute measure, 13*3. The Antarctic Expe-
dition of Sir James Ross has taught us, that the difference of
force between the two foci of the southern hemisphere is
probably less than between the two foci of the northern
hemisphere; and that the intensity of the force is itself greater
at each of the southern foci than at either of the northern
foci. The intensity of the force at the stronger southern
focus (lat. 64° S., long. 137° 30' E.) is in the relative
scale at least 2*06, (101) in the absolute scale 15'60 : at the
weaker southern focus,(102) lat. 60° S., long. ? 125° W.,
still, according to Sir James Ross, it is in the relative scale
1-96, and in the absolute 14' 90. The greater or less
distance apart of the two foci in the same hemisphere is
recognised as an important element of their individual
strength and of the whole distribution of the magnetic force
at the surface of the earth. " Although the foci of the
southern hemisphere present a strikingly greater magnetic
force (in absolute measure 15*60 and 14*90) than those of

H 2

100 rEURESTRIAL

the northern hemisphere (14*21 and 13*30), yet the mag-
netic force of the one hemisphere is not to be regarded as
greater on the whole than that of the other."

It is quite otherwise, however, if we divide the earth into
an eastern and a western hemisphere by a great circle
formed by the meridians of 100° and 280° E. long., so
that the eastern (and the more continental) hemisphere shall
comprehend South America, the Atlantic Ocean, Europe,
Africa, and Asia almost to Lake Baikal ; and the western
(which is the more oceanic and insular) shall contain
almost the whole of North America, the Pacific Ocean,
New Holland, and the eastern part of Asia. These meri-
dians are, the one about 4° west of Singapore, the other
13° west of Cape Horn; the latter being also the meridian
of Guayaquil. In this division all four foci of maximum
magnetic lorce, and even the two magnetic poles themselves,
all belong to the western and more oceanic hemisphere. (103)

Adolph Erman's important observation of the least or
minimum magnetic force in the Atlantic Ocean, on the
east side of the Province of Espiritu Santo in Brazil (lat.
20° S., long. 35° 02' W.), has been already referred to in
Vol. i.(10*). He found in the relative scale 0-706 ; in the
absolute, 5 "35. This region of weakest intensity was also
crossed twice by the Antarctic Expedition of Sir James
lloss,(105) between latitudes 19° and 21° S., and by Lieute-
nant Sulivan and Mr. Dunlop, on the passage from England
to the Falkland Islands. (106) Sabine has represented the
curve of least magnetic force [corresponding to the year
1840], which Eoss terms "the equator of least intensity/'
on the isodynamic map of the whole Atlantic Ocean, draw-
ing it from shore to shore. It cuts the west coast of Africa

TERRESTRIAL MAGNETISM. 101

at the Portuguese colony of Mossamedes (lat. 15° S.),
has its concave summit in the middle of the ocean in long.
18° W., and touches the Brazilian coast in 20° S. lat.
Future observations will show more clearly whether there
may not also be a region of comparatively weak magnetic
force (0-97 of the arbitrary scale) to the north of the
equator in 10° to 12° N. lat., and about 20° to the east of
the Philippines.

According to the data at present possessed, I believe I
have little to alter in the ratio given in Vol. I. of the weakest
to the strongest terrestrial magnetic force. It falls between
1 : 2J and 1 : 3, but nearest to the latter; some
diversity of statement being caused by alterations having
been rather arbitrarily made, sometimes in the minima only,
and sometimes in both minima and maxima at once.(107)
Sabine(108) has the great merit of having first called atten-
tion to the importance of the " dynamic equator," (the curve
of least intensity of the magnetic force). " This curve, which
undulates in its progress round the globe, connects the
points in each geographical meridian at which the earth's
magnetic force is least, the force increasing everywhere in
receding from it on either side towards the higher latitudes
of the two hemispheres. It marks the physical separation
between the two magnetic hemispheres more decidedly than
does the line of no dip, on which the direction of the
magnetic force is perpendicular to the direction of gravity.
For the theory of magnetism all that relates to the force
has a more immediate bearing than what relates to the
direction of the needle, either in the horizontal or vertical
plane ; those planes, although necessarily used by us both
in observation and discussion, not having in themselves
any direct relation to magnetism. The inflexions of the

102 TERRESTRIAL MAGNETISM,

dynamic equator are complex, because they depend on forces
which have four points or foci of greatest intensity which
are unsymmetrically distributed on the surface of the globe,
and are of unequal strength. The most remarkable of these
inflections is the great convexity towards the southern pole,
in the Atlantic Ocean, between the coast of Brazil and the
Cape of Good Hope/'

Does the intensity of the terrestrial magnetic force
diminish sensibly at accessible distances above the surface
of the earth ? or increase sensibly in descending below that
surface ? The problem which these questions propose for
solution is an exceedingly complicated one, in so far as it
has to be determined by observations made above or below
the ordinary level of the earth's surface. In attempting to
solve it by means of mountain journeys or descents into mines,
the facts that the upper and lower stations of observation are
seldom situated vertically in respect to each other, — that the
observations are frequently influenced by the character of the
rocks and the presence perhaps of hidden mineral veins, — and
that the fluctuations, regular or irregular, of the magnetic
force itself, where the observations are not strictly simulta-
neous, require to be allowed for ; — are all circumstances which
are likely to affect the results(108), and to cause to be ascribed
to differences of height or depth what may be by no means
due to them. Among the numerous mines which I have
visited, descending to considerable depths, in Europe, Peru,
Mexico, and Siberia, I have never found localities which were
suited to inspire me with confidence in reference to this ques-
tion :(109) it must also be remembered, that considering the
normal plane to be that of the level of the sea (viewed as the
mean general surface of the terrestrial spheroid), many mines,
thouorh offeriner considerable denths below the surface of tlip.

TERRESTRIAL MAGNETISM. 103

ground, never can reach the sea-level. Thus the works in
the Joachimsthal, in Bohemia, have reached 2000 feet of
absolute descent, but the lowest part is still 250 feet above
the sea.(110) Aerostatic ascents offer very different and far
more favourable conditions. Gay-Lussac ascended to 21600
French feet above Paris, a height eleven times greater than
the greatest depth attained in Europe, (speaking even merely
of depth beneath the surface of the ground, without
reference to the level of the sea). The results of my own
mountain observations between the years 1799 and 1806
led me on the whole to regard a decrease of force in ascend-
ing as probable, although, (perhaps from the perturbing
causes above alluded to), several amongst them gave indica-
tions of a contrary character. I have placed together in a
note,(m) data obtained by me in the course of 125 determi-
nations of magnetic force in the chain of the Andes, in the
Alps, and in Italy and Germany. My observations extended
from the level of the sea to the limits of perpetual snow, and
to a height of 14960 French, or 15944 English, feet; but
the greatest elevations did not give the most assured results.
The most satisfactory among them are afforded by the steep
declivity of the Silla de Caracas, 8100 French feet above
the closely-adjacent coast of La Guayra; the Santuario de
Ntra Sra de Guadelupe, on the summit of a limestone cliff
rising immediately above the town of Bogota, giving a
difference of elevation of 2000 feet ; and the volcano of
Purace, 8200 French feet above the Plaza Mayor of the
town of Popayan. Kupffer(112) in the Caucasus, Forbes in
many parts of Europe, Laugier and Mauvais on the Canigou,
Bravais and Martins on the Faulhorn, and in their adven-
turous sojourn very near the summit of Mont Blanc, have

104

TERRESTRIAL MAGNETISM.

indeed all remarked a decrease of magnetic force with increas-
ing elevation. It would seem, from the general discussion by
Bravais, as if the decrease were greater in the Pyrenees than
in the Alps.(113)

Quetelet's entirely opposite results, obtained in a journey
from Geneva to the Col de Balme and* the great St. Ber-
nard, render it doubly desirable that in order to obtain a
finally decisive reply to so important a question, recourse
should be had to the only effectual means, i. e. balloon
ascents, in which the surface of the earth is quitted alto-
gether, (a step taken so long ago as 1804, by Gay-Lussac,
first conjointly with Biot on the 24th of August, and then
alone on the 16th of September), and that a consecu-
tive series of experiments should be thus made. The time
of vibration of needles carried in aeronautic ascents to
elevations of eighteen or twenty thousand feet and upwards,
can only be made to afford just inferences respecting the
degree of the earth's magnetic force propagated through the
free atmosphere, when the temperature correction of the
needles employed is determined with great precision both
before and after the ascent. The neglect of any such correction
caused it to be erroneously inferred from Gay-Lussac's
experiments that the magnetic force remains unaltered to an
elevation of 21600 feet;(114) whereas the experiment really
went to indicate the decrease of the force, since the colder
temperature of the higher region must be supposed to have
acted in accelerating the vibrations of the needle. (115)
Neither ought Faraday's brilliant discovery of the paramag-
netic force of oxygen to be left out of view in considering
the subject before us. That great physicist calls attention
to the consideration, that the cause of the diminution of the

TERRESTRIAL MAGNETISM. 105

magnetic force in the higher strata of the atmosphere should
not be looked for solely in the increased distance from the
source of the force, the solid terrestrial globe, — but may
with n» less probability be sought in the exceedingly rarefied
state of the air at those elevations, whereby the quantity of
oxygen contained in a cubic foot of atmospheric air is much
less than at the surface of the earth. It appears to me that
we should not be justified in extending this supposition
further, than to say that the diminishing paramagnetic
property of the oxygen with increased rarefaction of the
air, at increased elevations, may be a concurrent modifying
cause of the phenomenon in question. Modifications of
temperature and of density, by the action of ascending
aerial currents, will, again, modify the measure of this con-
current action ;(116) and disturbing influences thus assuming
a variable and local character, will act in the atmosphere as
do the different kinds of rock at the surface of the earth.
At every cheering step in advance towards the better know-
ledge of the constituents of the gaseous envelope of our
planet, and their physical properties, we become aware of
fresh liabilities to error amidst the varying concurrent action
of forces, and are admonished thereby of the need of still
greater caution in arriving at conclusions.

The intensity of the earth's magnetic force measured at
determinate points of the earth's surface, has (like all the
other phenomena of terrestrial magnetism) its hourly, and
also its secular variations. The former were distinctly re-
cognised in Parry's third voyage, by that distinguished
navigator, and by Lieutenant Poster, in 1825, at Port
Bowen. In the middle latitudes the increase of the inten-
sity of the magnetic force from the morning to the evening

106 TERRESTRIAL MAGNETISM.

had been the subject of very careful examination by Chris-
tie^117) Arago, Hansteen, Gauss, and Kupffer. But as,
notwithstanding the great improvements in the construction
of dipping-needles, their time of vibration in the vertical
plane could not be ascertained with the same accuracy as
that of needles vibrating in the horizontal plane, a know-
ledge of the horary variations of the total force could not
be obtained from the latter, without a more exact knowledge
of the horary variations of the inclination than could be
gained by the instrumental means in use before the existence
of the magnetic observatories. The establishment of
magnetic stations in the northern and southern hemi-
spheres has since afforded the great advantage of supply-
ing results at once the most numerous and the best
assured, with magnetometers measuring the horary varia-
tions of the horizontal and vertical components of the
force, and thus enabling their theoretical equivalents,
the horary variations of the inclination and of the total force,
to be learnt. It will be sufficient here to consider two
of these stations, (ll8) both extra-tropical, and situated
at nearly equal distances on either side of the equator ; viz.
Toronto in Canada, in 43° 32' N. lat., and Hobarton in
Van Diemen Island in 42° 53' S. lat.; their difference of
longitude being- about 15 hours. In the simultaneous
system of hourly observations, the observations of the
winter months of the one station are made during the
summer of the other, and the greater part of the night
observations at the one correspond in like manner to the
day observations at the other. The declination is at
Toronto 1° 33' W., and at Hobarton 9° 57' E. ; the incli-
nation and force at the two stations are very similar, the

TERRESTRIAL MAGNETISM. 107

dip of the north end of the needle at Toronto being 75° 15',
and that of the south end at Hobarton 70° 34', — and the
total force in absolute measure 13'90 at Toronto, and 13'56
at Hobarton. Of these two so well-chosen stations, the
Canadian shows, according to Sabine's investigation, (ll9)
four, and that of Yan Diemen Island only two, turning-
points in the diurnal variation of the total force. At Toronto
there appears to be a principal maximum at 5 h., and a prin-
cipal minimum between 15 h. and 16 h. ; with a weaker
secondary maximum varying in different months from 18 h.
to 20 h., and a weaker secondary minimum at 22 h. or 23 h.
At Hobarton, on the other hand, the intensity of the force
follows a single progression from a maximum between 5 h.
and 6 h. to a minimum between 20 h. and 21 h. ; although
the inclination at that station has, as at Toronto, four
turning-points. (]2°) By combining the variations of incli-
nation with those of the horizontal force, Sabine has found
that at both stations the total force is greatest from October
to February, and least from April to August ; October to
February being months of winter at Toronto, and of summer
at Hobarton ; and April to August being months of summer
at Toronto and winter at Hobarton. He thence infers that
it is not to differences of temperature that we should ascribe
these variations, and suggests that the increase of the total
magnetic force in both hemispheres during the months
when the Sun is in the southern signs, may be caused by the
greater proximity of the Earth in that portion of her orbit to
the Sun acting as a magnetic body.(121) At Hobarton, the
intensity of the force is in the summer of that station in
absolute measure 13'574, and in winter 13'54<3. The
secular change of the total magnetic force rests as yet on

108 TERRESTRIAL MAGNETISM.

the observations of very few years. The comparison of my
results with those of Rudberg, in the years 1806 and 1832,
would indicate for Berlin a small decrease. (122)

Inclination.

We are indebted for the knowledge we possess of the
geographical position of both the magnetic poles where the
Inclination is 90°, to the observations and scientific activity
of one and the same adventurous navigator, Sir James Ross :

O •*

in the north, in the second expedition (123) of his uncle Sir
John Ross, (1829 to 1833), arid in the south in the Antarctic
expedition commanded by himself (1 839 — 1843). According
to his observations, the northern magnetic pole (in 70° 5' N.
lat., 96° 43' W. long.) is five degrees of latitude further from
the pole of the earth than the southern (75° 5' S. lat.,
154° 10' E. long.) ; and the difference of longitude between
the two magnetic poles is 109°. The northern pole is situ-
ated on the large island of Boothia Felix (very near the
American continent), a part of which had been previously
called by Captain Parry, North Somerset. The observations
of Sir James Ross place the pole, at the date referred to, at a
short distance from the western coast of Boothia Felix, not
far from Cape Adelaide, which projects between King
William's Sea and Victoria Strait. (124) The south magnetic
pole has not been, like the northern one, directly reached.
On the 17th of February, 184i, Sir James Ross in the
'Erebus/ attained 76° 12' S. lat. in the meridian of
163° 02' E. long., but the magnetic inclination did not
exceed 88° 40', so that the south magnetic pole was supposed
to be still 160 English nautical miles distant. (125) Numerous
and very exact observations of declination, (determining the

TERRESTRIAL MAGNETISM. 109."

intersection of the magnetic meridians), render it very
probable that the present position of the south magnetic
pole is in the interior of the great Antarctic land called
South Victoria, west of the Albert Mountains, of which the
active volcano Mount Erebus, rising to the height of more
than 12000 feet, forms a part.

The situation, and the alterations in the form of the
magnetic equator, or line on which the dip is 0, have been
already spoken of by me in the " Description of Nature" in
Yol. i. (S. 190—192 and 431 ; Engl. p. 172—174, and
411). The earliest determination of the African node
(the intersection of the geographical and rnagnetical
equators) was by Sabine, at the commencement of his
Pendulum Expedition in 1822 (126) : at a later period, 1840,
the same savant, by combining the observations of Duperrey,
Allen, Dunlop, and Sulivan, formed a map of the magnetic
equator (127), from the west coast of Africa, (4° N. lat.,
9° 32' E. long.) through the Atlantic Ocean and Brazil
(16° S. lat. between Porto Seguro and Rio Grande), to the
point where I had found north dip change to south, on the
Cordilleras, not far from the Pacific. The African node,
or point of intersection of the two equators, was, in 1837, in
3° 02' E. long. ; in 1825 it had been in 6'57' E. long. The
secular movement of the node, in receding to the westward
from the lofty basaltic island of St. Thomas, had there-:
fore been at the annual rate of rather less than half a degree,
thus causing the line of no dip to impinge on the African
coast at a progressively more northern point ; whilst at the
same time it descended more to the south on the Brazilian
coast. The convex summit of the magnetic equator con-
tinues to be directed towards the south, its maximum

110 TERRESTRIAL MAGNETISM.

distance from the geographical equator in the Atlantic
Ocean being 16°. In the interior of South America, in the
Terra incognita of Matto Grosso, between the great rivers
Xingu, Madera, and Ucayale, there is an entire absence of
observations until the chain of the Andes is reached, where,
68 miles east of the Pacific, between Montan, Micuipampa,
and Caxamarca, I determined astronomically the place of
the magnetic equator, (7° V S. lat., 78° 46' W. long.),
which is there in course of ascending to the north- west (lii8)
The most complete investigation which we possess
towards a knowledge of the whole course of the magnetic
equator, is that made by my friend Duperrey, for the years
1823 — 1825. In the course of his voyage of circumnavi-
gation, he crossed the magnetic equator six times, and has
been enabled to lay it down from his own observations for
220 degrees of longitude. (129) The two nodes are situated,
according to Duperrey's map of the magnetic equator, one in
about 6° E. long, in the Atlantic Ocean, the other in 177^°
E.long., in the Pacific, between the meridians of the Viti and
Gilbert Islands. After quitting the west coast of the South
American continent, probably between Punta de la Aguja and
Payta, the magnetic equator continues, in its prolongation
westward, to approach the geographical equator, until, in the
meridian of the Mendafia group of islands, the two equators
are only two degrees apart. (13°) Ten degrees further to the
west, in the meridian of the western part of the Paumotu
Islands, or Low Archipelago, in 154° E. long., Captain
Wilkes, in 1840, found a similar distance of fully two
degrees of latitude between the geographic and magnetic
equators. (1S1) The intersection or node in the Pacific is
not 180° from the node in the Atlantic, being situated, not

TERRESTRIAL MAGNETISM. 11 J

in 174° W. long., but in the meridian of the Viti group,
in about 177^° E. or 182i° W. long., Therefore, in
proceeding to the west from the coast of Africa, we find the
distance between the two nodes Si° greater than the earth's
semi-circumference, — a proof that the curve in question is
not a great circle.

According to the excellent and widely-extended determi-
nations of Captain Elliot (1846—1849), which between the
meridians of Batavia and Ceylon are remarkably accordant
with those of Jules de Blosseville (see p. 67 of the present
volume), the line of no inclination passes across the north
of Borneo, and running in an almost exactly east and west
direction, touches the north point of Ceylon (9° 45' N. lat.)
The curve of least total force is in this part of the world
almost parallel with that of the magnetic equator. (132) The
latter enters the east coast of Africa south of Cape Gardafui :
this important point has been determined with great exact-
ness by Rochet d'Hericourt in his second Abyssinian expe-
dition (1842 — 1845), and by the able discussion of that
traveller's magnetic observations^133) It is situated south
of Gaubade, between Angolola and Angobar (the principal
town of the kingdom of Shoa), in 10° 7' N. lat. and
41° 13' E. long. The course of the magnetic equator
through the interior of Africa, from Angobar to the Bight
of Biafra, is as entirely unexamined as is the portion of the
same line which passes through that part of the interior of
South America which is east of the chain of the Andes and
south of the geographical equator. These two continental
spaces are of about equal extent in an east and west direc-
tion, and taken together occupy 80 degrees of longitude, or
nearly a quarter of the earth's circumference, in which there

112 TERRESTRIAL MAGNETISM.

is thus as yet an entire absence of magnetic determinations.
My own observations of inclination and force through the
whole interior of South America (from Cumana to Rio
Negro, as well as from Cartagena de Indias to Quito) were
confined to the tropical zone north of the geographical
equator ; those in the southern hemisphere, from Quito to
Lima, only extended over the narrow district adjacent to the
western coast.

The movement of translation of the African node to the
westward from 1825 to 1837, noticed in a preceding page,
is confirmed, on the east coast of Africa, by a comparison of
the inclination observed by Panton, in 1776, with Rochet
d'Hericourt's observations. The last-mentioned traveller
found the line of no dip much nearer the Straits of Bab-el-
Mandeb, i. e. 1° south of the island of Socotora in 8° 40'
N. lat. Thus there would appear to have been in 49 years
only an alteration of 1° 27' in latitude : the alteration in
longitude in the same interval of time, by the movement of
the node as estimated by Arago and Duperrey, would
amount to 10° to the westward. The direction of the
movement due to secular change on the eastern side
of Africa has therefore been quite the same as on
the western, but the quantity or amount of the move-
ment still requires to be determined by more exact
results.

The periodicity of the variations in the magnetic inclina-
tion, in reference to hours and seasons, the existence of
which has been already remarked, has only been established
definitively, and in its entire character, within the last twelve
years, or since the formation of the British magnetic stations
in the two hemispheres. Arago, to whom so much is due

TERRESTRIAL MAGNETISM. 113

in reference to magnetism, had indeed recognised in the
autumn of 1827 that " the inclination is greater at 9 A.M.
than at 6 P.M. ; while the intensity of the magnetic force, as
measured by the vibrations of a horizontal needle, attains its
minimum at the former, and its maximum at the latter, of
these two epochs/' (134) The whole diurnal march of the
inclination has now been solidly established by means of
many thousand regularly continued hourly observations at
the British magnetic observatories since 1840, and by their
laborious discussion. This is the place for bringing together
the obtained facts as the foundations of a general theory of
terrestrial magnetism. Before doing so, it is desirable to
remark that, in considering the periodic fluctuations of the
three elements of terrestrial magnetism in their entire
character, we ought, with Sabine, to distinguish in the
" turning hours/' or hours of maxima or minima, between
two greater, and therefore important, extremes ; and other
intervening minor (though for the most part not less
regular) fluctuations. The recurring movements of the
inclination and declination needles present to us, then, as
do the variations of the intensity of the total force, principal
and secondary maxima and minima ; in the most usual cases
both principal and secondary, forming a double progression
with four turning hours, but less commonly a single maximum
and minimum only, forming a simple progression with but
two turning hours. The march of the total force in Van
Diemen Island, for example, is of the latter description, while
the inclination at the same station follows a double progres-
sion ; and at Toronto, in Canada, a place whose position in
the northern hemisphere corresponds almost exactly to that of
Hobarton in the southern hemisphere, the diurnal march of
VOL. IT.

1 14 TERRESTRIAL MAGNETISM.

the total force is a double progression, whilst that of the incli-
nation is a double progression only from October to March,
and a single progression from April to September. (l35) At
the Cape of Good Hope the inclination has only a single
maximum and single minimum, whilst the total force has a
double progression, the principal minimum occurring at the
same hour as the minimum of inclination.

We must also distinguish between results obtained by a
series of observations with a dipping-needle at certain hours
of the forenoon compared with a similar series at certain
hours in the afternoon (which can at most only give the
difference in the amount of the inclination at those two
periods of the twenty-four hours), and results obtained by
hourly observations of the horizontal and vertical force
magnetometers, which give the horary variations of the incli-
nation and total force for every hour. Amongst the horary
variations of the inclination obtained by either of these
methods may be cited the following : — -

I. In the Northern Hemisphere.

Greenwich. Prom observations with a dipping-needle
three hours before and three hours after noon, the north dip
was found to be greater at 9 A.M. than at 3 P.M. The
difference in 184? was 0''7. In four years out of five, the
dip was higher at 9 A.M. than at 3 P.M., but in one year
(1845) the reverse appeared, the dip being greater by
T-3 at 3 P.M. than at 9 A.M.

Paris. Erom observations with a dipping-needle at 9 A.M.
and 6 P.M., the mean north dip appeared to be greatest at

9 A.M.

TEURESTRIAL MAGNETISM. 115

Petersburg. From observations with a dipping-needle
at 8 A.M. and 10 P.M., the mean north dip appeared to be
greatest at 8 A.M.

Toronto, in Canada. Prom hourly observations during
5J years with horizontal and vertical force magnetometers,
a principal maximum is found in all months of the year
about the hour of 4 P.M., occurring, however, somewhat
earlier from April to September than from October to
March ; and a principal minimum about 1 0 A.M. or 11 A.M.,
occurring also earlier from April to September than from
October to March. The progression from the maximum at
10 or 11 A.M. to the minimum at 4 P.M. is continuous and
rapid. From April to September the inclination increases,
with occasional very slight interruptions, from the minimum
at 4} P.M. to the maximum at 10 A.M. At this season,
therefore, the horary variation scarcely differs from a single
progression, the decrease taking place in the six hours from
10 A.M. to 4 P.M., and the increase, more slowly, in the
remaining eighteen hours. In the opposite season, i. e. from
October to March, a secondary maximum shows itself at
from midnight to 2 A.M., and a secondary minimum at
about 6 A.M. (Sabine, Toronto, Yol. ii. p. Ixx.) The north
dip is greater in the six months when the Sun is in the
southern signs (75° 17''S4) than in the six months when
the Sun is in the northern signs (7 5° 1 6'*5 7) . The intensity
of the total force is also greater by about two-thousandth
parts of its whole amount in December and January, when
the Earth is nearest to the Sun, than in June and July,
when the Earth is most distant from the Sun. (Toronto
Obs. Vol. ii. pp. Ixxxvii. and xcii. and xciii.)

i 2

116 TEUftESTRTAL MAGNETISM.

II. In the Southern Hemisphere.

Holarton, Van Diemen Island. From hourly observa-
tions during six years with horizontal and vertical force
magnetometers, the principal maximum (of south dip) occurs
at 11 J A.M.; the principal minimum at 6 A.M. ;" a secondary
maximum at 10 P.M.; and a secondary minimum at 5 P.M.
(Sabine, Hobarton, Vol. i. p. Ixvii.) The south dip is greater
in the six months when the Sun is in the southern signs
( — 70° 36'-60) than when in the northern signs
( - 70° 35'-42). The intensity of the total force is also
greater at Hobarton from December to February than
from June to August. (Hobarton Obs. Vol. ii. p. xlvi.)

Cape of Good Hope. From hourly observations during
4 1 years with horizontal and vertical force magnetometers,
a single progression is found. Maximum at 8 h. 34 m. A.M. ;
minimum at 0° '*34 P.M. ; with a very small intervening
fluctuation between 7 A.M. and 9 A.M.

In comparing the two stations of Toronto and Hobarton,
situated in corresponding latitudes on either side of the
equator, we notice remarkable correspondencies in respect
to the turning hours ; thus

10 to 11J A.M. is the epoch of principal minimum at
Toronto, and of principal maximum at Hobarton.

4 P.M. is the epoch of principal maximum at Toronto,
and 5 P.M. of secondary minimum at Hobarton.

6 A.M. is the epoch of principal minimum at Hobarton,
and of secondary minimum at Toronto ; and from

TEKRESTRIAL MAGNETISM. 117

10 P.M. to 2 A.M. a secondary maximum occurs at both

stations.

The four turning hours of the inclination at Toronto
are almost exactly reproduced at Hobarton, only the sig-
nification is altered. This complex action is very deserving
of attention ; as is also the comparison of the two stations
in respect to the sequence of the turning hours of the varia-
tions of the inclination and of the total force. (136)

The periods of the inclination at the Cape of Good Hope
do not agree either with Hobarton, which is in the same
hemisphere, or with any of the northern stations which have
been referred to. The minimum of inclination even takes
place at an hour when the inclination at Hobarton has
almost reached its maximum.

The determination of the secular change of the inclination
requires observations of equal and satisfactory accuracy,
repeated so as to include long intervals of time. We do
not, for example, find that we can go back to Cook's
voyages with the desired degree of certainty ; for although
in his third voyage the poles of the dipping-needle were
always reversed, yet differences from forty to fifty-four
minutes occur between that great navigator's observations
in the Pacific and those of Bayley, — attributable probably to
the then very imperfect construction of the needles, and
especially to their want of free movement. For London,
we are reluctant to go back beyond Sabine's observations in
August 1821, which, compared with the excellent deter-
mination by James Ross, Sabine, Johnson, and Fox, in
1838, gave for that interval an annual rate of decrease of
2'*73 ; in near accordance with which, Lloyd, with equally

118 TERRESTRIAL MAGNETISM.

exact instruments, but in a shorter interval of time, found for
Dublin 2''38 annual decrease. (137) In Paris the inclination
is also diminishing, and more rapidly than in London. The
very ingenious methods of determining the dip devised by
Coulomb were not practically successful in the hands of
their inventor, for he was led by them to erroneous
results. The first observation with a good instrument of
Lenoir^s was made in 1798 at the Observatory of Paris,
when, conjointly with Chevalier Borda, I found by several
repetitions 69° 51'*0 ; in 1810, I found, conjointly with
Arago, 68° 50''2; and in 1826, with Mathieu, 67° 56''7.
In 1841 Arago found 67°9''0; and in 1851 Laugier and
Mauvais found 66° 35' : all these determinations were
made by the same method, and with similar instruments.
The entire period, exceeding half a century, (1798 to
1851) gives a mean annual decrease of inclination at Paris
of 3/f69, being for the several intervals :

i from 1798 — 1810 at the rate of 5'-08
1810 — 1826 „ 3'-37

1826 — 1841 „ 3'-] 3

1841 — 1851 „ 3'-40

The rate of decrease underwent a remarkable retardation
between the first and second interval, but the change was a
gradual one, for a very careful observation of Gay-Lussac's,
made in 1806 on his return from Berlin, where he had
accompanied me after our return from our Italian journey,
gave the inclination 69° 12', corresponding to a rate of
decrease of 4*87. The nearer the node of the magnetic
equator, in its secular movement from east to west, ap-

TERRESTRIAL MAGNETISM. 119

preaches the meridian of Paris, the more the rate of decrease
appears to slacken, the retardation being in the course of
half a century from 5''08 to 3''40. In a memoir which I
presented to the Berlin Academy in April 1829, a short
time before my Siberian expedition, (138) I collected and
compared the observations made by me, — I think I may
venture to say, always with equal care. Sabine measured
the inclination and the force at the Havannah fully twenty-
five years after my observations there, thus giving for that
tropical station the variation of two important magnetic
elements for a considerable interval. Hansteen (1831) has
examined the annual variation of the inclination in either

hemisphere in a very meritorious and more comprehensive
work (139) than mine.

While the observations of Sir Edward Belcher in 1838,
compared with mine in 1803, indicate considerable changes
of inclination along the west coast of America, between
Lima, Guayaquil, and Acapulco (vide page 77) (the longer
the interval, the greater the value of the results), the secular
change in other parts of the Pacific appears to have been re-
markably small. At Otaheite, Bayley found, in 1773, 29° 43';
Fitz Boy, in 1835, 30° 14'; Belcher, in 1840, 30° IT, so
that in sixty-seven years the mean annual increase would
hardly have been 0'-51(14°) In Northern Asia, also, a
very careful observer, Mr. Sawelieff, twenty-two years after
my visit to those regions, found on a journey which he
made from Kasan to the shores of the Caspian, that very
unequal changes had taken place in the inclination on the
north and south sides of the parallel of 50° of geographical
latitude. (141)

120 TERRESTRIAL MAGNETISM.

Humboldt. Sawelieff.

1829. 1851.

Kasan. . . . 68° 26'-7 68° 30'-8

Saratoff . . . 64° 40'-9 64° 48'-7

Sarepta . . . 62° 15''9 62° 39'-6

Astrachan . . 59° 58''3 60° 27'-9

Tor the Cape of Good Hope we possess a very long, and
if we do not go farther back than from Sir James Ross and
Du Petit Thouars in 1840, to "Vancouver in 1791, a very
satisfactory comparison of observations of inclination, com-
prising an interval of almost fifty years. (142)

The question whether the height of the station has any
unequivocally discernible influence on the magnetic inclina-
tion and force, (l43) was the subject of careful examination
by me in my journeys in the Andes, the Ural, and the
Altai Mountains. I have already remarked, in the section
on the Force, how few unfortunately are the localities which
are suited to throw any certain light upon this inquiry, as
the points compared ought to be so near to each other as
to avoid all suspicion that the magnetic differences may be
due, not to the difference of elevation, but to the inflections
of the isodynamic and isoclinal curves, or to diversity in
the character of the rocks. I will confine myself to the
four principal results of the inclination in respect to which
I thought at the time, and on the spot, that they indicate,
with greater certainty than the force observations, an
influence of elevation in diminishing the inclination of the
needle.

On the Silla de Caracas, which rises almost perpendicu-
larly 8100 Trench feet above the sea-coast of La Guayra,

TEERESTIUAL MAGNETISM.

south of the coast and north of the town of Caracas,
incl. 41°-90 : La Guayra, elev. 10 feet; incl. 42°'20 ; town
of Caracas, elev. 2484 French feet; incl. 42°'95. (Hum-
boldt, Yoy. aux Kegions equinox. T. i. p. 612.)

Santa Fe de Bogota: Elev. 8196 French feet; incl.
27°'15 : Chapel de Nuestra Senora de Guadalupe, on a
precipice above the town, elev. 10128 French feet; incl.
26°-80.

Popayan : Elev. 5466 French feet ; incl. 23°'25 : moun-
tain village of Purace, on the declivity of the volcano,
elev. 8136 French feet; incl. 21°'80 : summit of the vol-
cano of Purace, elev. 13650 French feet; incl. 20°'30.

Quito: Elev. 8952 French feet; incl. 14°'S5 : San
Antonio de Lulumbamba, where the geographical equator
passes through the hot valley ; elev* of the bottom of the
valley, 7650 French feet; incl. 16°'02. All the above
inclinations are given, as observed, in centesimal degrees, as
their comparative value is the only point under consideration.

On account of the too great horizontal distances between
the stations, and the influence of adjacent rocks, I do not-
lay aoy stress on my European observations : as, for example,
at the Hospice on St. Gothard (6650 French feet), incl.
66° 12/,compared to Airolo (3502 French feet),incl. 66° 54',
and Altorf, incl. 66° 55'; or (which seem to tell the other
way) Lans le Bourg, incl. 66° 9'; Hospice on Mount Cenis
(6358 French feet), incl. 66° 22'; arid Turin (707 French
feet), incl. 66° 3' ; or at Naples, Portici, and the crater of
"Vesuvius ; or in Bohemia, the summit of the great Mili-
schauer (a Phonolite !), incl. 67°53'; Toplitz, incl. 67°19''5;
and Prague, incl. 66° 47''6.(144) In 1844 Bravais, in
company with Martins and Lepileur, made a series of

122 TERRESTRIAL MAGNETISM.

excellent comparative horizontal force observations (pub-
lished in great detail) at 35 stations, including the
summits of Mont Blanc (14809 Trench feet), the Great
St. Bernard (7848 French feet), and the Faulhorn (8175
French feet) ; contemporaneously with which were observa-
tions of the inclination on the Grand Plateau of Mont
Blanc (12097 French feet), and at Chamounix (3201
French feet). If the comparison of these results would
indicate a diminution of inclination with increased ele-
vation, observations on the Faulhorn and at Brienz
(1754 French feet) on the other hand, would give the
contrary indication. Thus no satisfactory solution of the
problem was obtained by either class of experiment.
(Bravais sur Fintensite du Magnetisme terrestre en France,
en Suisse et en Savoie, in the Ann ales de Chimie et de
Physique, 3eme Serie, T. xviii. 1846, p. 225.) In a
manuscript of Borda's on the subject of his expedition to
the Canaries in 1776, (preserved in the Depot de la Marine
at Paris, and for the communication of which I am indebted
to Admiral Rosily), I have found the proof that Borda made
the first attempt to examine the influence of height on this
magnetic element. He found the inclination on the summit
of the Peak of Teneriffe 1° 15' greater than in the port of
Santa Cruz, — a difference, no doubt, in great part at least,
the consequence of the local attraction of the lavas, similar to
that which I have myself often observed on American
volcanoes and on Vesuvius. (Humboldt, Yoy. aux Beg.
equinox. T. i. pp. 116, 277, and 288.)

With the view of trying the analogous question of the
influence of depth below the earth's surface, being at Frei-
berg in July 1828, I made observations of the inclination

TERRESTRIAL MAGNETISM. 123

with the greatest care of which I am capable, and always
reversing the poles of the needles, in a mine in which the
strictest examination could detect no influence of the rock
(gneiss) upon a magnetic needle. The depth below the
surface was 802 Trench feet. The difference of the subterra-
nean inclination from that observed at a point immediately
above it is, indeed, only 2'- 06, but from the precautions
taken in the whole proceeding, the results of the several
needles which are stated in a note,(145) induce me to
believe that the inclination is really greater below than at the
surface. It is very desirable to repeat similar trials with
care, in favourable localities and where it can be ascertained
that the strata are free from all local influence, in mines of
greater depth, — as that of Yalenciana, near Guanaxuato in
Mexico, 3582 French feet, — in English coal mines more
than 1800 feet deep, — or in the now ruined Eselschacht,(146)
at Kuttenberg in Bohemia, 3545 French feet in vertical

After a violent earthquake at Cumana, on the 4th of
November, 1799, I found the inclination was diminished
90 centesimal minutes, or nearly a whole degree. The
circumstances under which I obtained this result, of
which I have given elsewhere an exact relation, (li7) offer no
satisfactory reason for supposing an error. A short time
after landing at Cumana, I had found the inclination
43°*53 (Centesimal). The accident of having seen, a few
days before the earthquake, in an otherwise estimable
Spanish work, Mendoza's Tratado de Navegacion, T. ii.
p. 72, the erroneous opinion, that the diurnal and annual
variations of the inclination are greater than those of the

124 TERRESTRIAL MAGNETISM.

declination, had occasioned me to institute, in the harbour
of Cumana, a series of careful observations. "Prom the
1st to the 2nd of November the inclination continued with
great steadiness to show a mean amount of 43°'65. The
instrument remained untouched and properly levelled, in
the same place. On the 7th of November, three days after
the strong earthquake shocks, the instrument, after being
levelled afresh, gave 42°*75. The intensity of the magnetic
force, measured by vertical oscillations, was unaltered. I
hoped that the inclination would return, perhaps gradually,
to its former value ; but it remained the same. On return-
ing to Cumana in September 1800, after travelling by
river and land journeys on the Orinoco and Eio Negro
upwards of two thousand geographical miles, the same
dipping-needle of Borda's, which had accompanied me every-
where, gave the inclination 42°*80, — very nearly the same as
in the last observation before it left Cumana. As mechanical
agitations and electric shocks, by altering the molecular con-
dition in soft iron, elicit poles, so we may conceive it possible
that there may be a connection between the direction of
magnetic currents and that of earthquake shocks; but
although my attention was strongly directed towards a
phenomenon of the objective reality of which I had in
1799 no reason to doubt, yet, in the numerous earthquake
shocks which I experienced during the three years subse-
quently passed in South America, I never again met with a
sudden alteration of the magnetic inclination which I could
ascribe to such a cause, various as were the directions in
which the undulatory movement of the terrestrial strata
was propagated. A very accurate and experienced observer,

TERRESTillAL MAGNETISM. 125

Erman, also found, after an earthquake on Lake Baikal on
the 8th of March, 1828, no disturbance either in the amount
of the declination or in its periodic variation. (l48)

Declination..

I have already touched, in the earlier part of the present
volume, on the historical facts of the earliest recognition of
phenomena relating to the third element of terrestrial
magnetism, — the declination. In the 12th century of our
era, the Chinese were not only acquainted with the fact of
the deviation of a horizontal magnetic needle, suspended by
a cotton thread, from the geographical meridian, but they
also knew how to determine the amount of this deviation.
When afterwards, by the intercourse of the Chinese with
the Malays and Indians, and of these with the Arabs and
the Moorish pilots and navigators, the use of the mariners'
compass became common among the Genoese, Majorcans,
and Catalans in the Mediterranean, on the west coast of
Africa, and in the Northern Seas, indications of the varia-
tion (or declination) came to be introduced in nautical
charts of different parts of the ocean, even as early as
1436. (149) The geographical position of a line of no
variation, on which the needle pointed to the true north or
pole of the Earth's rotation, was determined by Columbus
on the 13th of September, 1492 : it even did not escape
him that the knowledge of the magnetic declination might
serve to determine the geographical longitude. I have
shown elsewhere, from his ship's journal, that on his second
voyage (April 1496), when uncertain about his ship's
reckoning, he sought to aid himself by observations of

] 26 TERRESTRIAL MAGNETISM.

declination. (15°) The horary variations or changes of decli-
nation were recognised as facts by Hellibrand and Tachard,
at Louvo, in Siam ; they were first observed, circumstan-
tially and almost satisfactorily, by Graham in 1722. Celsius
was the first who arranged concerted simultaneous measure-
ments of these variations by different observers at two
distant points. (l51)

Passing now to the phsenomena presented by the declina-
tion of the magnetic needle, we will consider them, first, in
reference to their variations according to the hours of the
day or night, the seasons of the year, and their mean state
in different years ; next in regard to the influence exercised
on those variations by the extraordinary, and yet periodic,
disturbances, and by the position of the places of obser-
vation to the north or south of the magnetic equator ; and
lastly, we will consider them according to the linear relations
of places on the earth's surface having equal, or (it may be)
no, declination. These linear relations are indeed that part
of the acquired knowledge which is the most important in
its direct and practical application to navigation ; but all
the general magnetic phsenomena (among which the extra-
ordinary disturbances, or magnetic storms, acting often
simultaneously at such great distances, are among the most
mysterious), are so intimately connected together, that, with
a view to the gradual completion of the mathematical
theory of terrestrial magnetism, we should neglect none of
them.

In the middle latitudes of the northern magnetic hemi-
sphere (dividing the earth at the magnetic equator), the
north end of the needle, «'. e. the end which points in those
latitudes towards the north, points more to the east at

TERRESTRIAL MAGNETISM. 127

8JA.M. (20 J astronomical reckoning) than at any other hour,
making then the nearest approach to the true or geographical
north at all places where the declination is westerly. From
that hour this end of the needle moves gradually westward
until 1} P.M., when it reaches its most westerly elongation.
This movement to the westward is general ; it takes place
in the same direction at all places in the middle latitudes of
the northern hemisphere, whether they have west declination,
as the whole of Europe, Pekin, Nertschinsk, and Toronto
in Canada ; or east declination, as Kasan, Sitka (in Russian
America), Washington, Marmato in New Granada, and
Payta on the coast of Peru.(152) Erom the above-mentioned
most westerly pointing at If h., the needle returns towards
the east during the afternoon and a portion of the night
until 12 h. or 13 h., — often making, however, a small pause
at about 6 h. In the night there is again a small movement
to the westward, after which the easterly march is resumed
until the most easterly pointing of the needle is attained,
which, as already stated, is at 20J h. astr., or 85 A.M. This
nocturnal period, which was formerly quite overlooked, — a
gradual and uninterrupted progression to the east from
If P.M. until the morning hour of 8 having been the
general belief, — strongly engaged my attention at Rome,
when I was occupied there with Gay-Lussac in examining
the horary variations of the declination with a Prony's
magnetic telescope. The needle being generally more
unquiet while the sun is below the horizon, this small
nocturnal movement to the west presents itself on that
account both less frequently and less distinctly to the
recognition of the observer. When it does show itself
distinctly, I have noticed it to be unaccompanied by any

128 TERRESTEIAL MAGNETISM.

agitation of the needle. Quite differently from what is the
case during what I have termed magnetic storms, the needle
in this small westerly night excursion travels quietly from
one scale division to another, just as it does in the well-
assured and strongly characterised day movement between
20£h. and If h. It is very noticeable, and well deserving
of attention, that when the needle exchanges its continuous
westerly movement for an easterly one, or vice-versd, it
does not remain for a time without altering its direction,
but (especially in the case of the 20 J h. to If h. period)
turns suddenly back. The small westerly movement
usually takes place between midnight and early morning;
but it has been noticed in Berlin, in the Freiberg subterra-
nean observations, at Greenwich, at Makerstoun in Scotland,
at Washington, and at Toronto, as early as between 10 and
11, or 11 and 12 hours.

The four movements of the declination -needle recognised
by me in 1805(153) are presented as the result of many
thousand two-hourly observations made at Greenwich in the
years 1845, 1846, and 1847, four turning hours being
there assigned as follows(154) : — Principal minimum of
westerly declination at 20 h. ; principal maximum at 2 h. ;
secondary minimum at 12h. or 14 h.; secondary maximum
at 14 h. or 16 h. (the declination being westerly). I must
here content myself with giving the mean hours, and calling
attention to the circumstance that in our northern zone the
hour of the principal morning minimum (20 h.) is not at
all altered by the earlier or later time of sunrise. During
two solstitial and three equinoctial periods, for each of
which Oltmanns and myself followed the march of the
horary variation of the declination for five or six days and as

TERRESTRIAL MAGNETISM. 129

many nights, I always found the easternmost turning-point
of the needle, in the summer as in the winter months,
between 19| h. and 20£ h. ; being very slightly, (155) if it
might be said to be at all, accelerated, by the earlier time ot
sunrise in the summer.

In the high northern latitudes, near or within the arctic
circle, the regularity of the horary variation has been as
yet but imperfectly made out, although we possess a number
of very exact observations. In Iceland, Lottin, in the
French scientific expedition of the 'Lilloise' in 1836, was
almost afraid (considering the local influence of rocks, and
the frequency of Auroras) to derive any determinate results
in respect to the turning hours, either from his own
extensive and laborious observations, or from the earlier
ones (1786) of the meritorious Lowenorn. On the whole, at
Reikiavik, in Iceland, lat. 64° 8', as well as at Godthaab on
the Greenland coast according to the missionary Genge's
observations, the minimum of westerly declination appears
to fall almost as in the middle latitudes, at 21 h. or 22 h.,
but the maximum not until 9 h. or 10 h. in the evening. (156)
Further to the north, at Hammerfest, in Finmarken, in
lat. 70° 40', Sabine found the march of the declination
tolerably regular, (157) and similar to that in the south of
Norway and in Germany; westerly minimum at 21 h.,
westerly maximum at 1J h.; but he found it very different
in Spitzbergen in lat. 79° 50 , where the corresponding
turning hours were 18 h. and 7J h. For the Arctic Islands
north of America, we have at Port Bowen, on the eastern
side of Prince Regent' s Inlet (lat. 73° 14'), from Captain
Parry's third voyage (1825), a fine series of five months'
consecutive observations by Lieuts. Foster and James Ross:

YOL. IV. K

130 TERRESTRIAL MAGNETISM.

but although the needle passed twice in the twenty-four hours
through the direction regarded as the mean magnetic
meridian of the place, — and although for fully two months
(April and May) no Aurora was seen, — yet the times of the
principal elongations varied from four to six hours ; nay, more,
the mean epochs of the maxima and minima of west declina-
tion from January to May were only one hour apart ! The.
amount of variation was so great that on some days the
declination varied from 14° to 6° and 7°; (within the
tropics the differences hardly reach as many minutes. (158)
Not only within the polar circle, but also within the tropics,
e. (/. at Bombay (lat. 18° 56'), there is great complexity in
the periods of the horary variation of the declination. They
there fall under two principal classes, being very different
from April to October and from October to December ; and
these again subdivide into two minor periods, which are
far from being well defined. (159)

European nations knew from their own experience nothing
respecting the direction of the magnetic needle in the
southern hemisphere until the second half of the 15th
century, when through the adventurous voyages of Diego
Cam and Martin Behaim, Bartholomew Diaz and Yasco
de Gama, some slight notice on the subject reached Europe.
The importance which, as we learn from their early writers,
was attached to the south end of the magnetic needle by the
Chinese (who, as well as the inhabitants of Corea and of
the islands of Japan, guided themselves at sea as well as
on land by the compass as early as the 3rd century of our
era), was ho doubt occasioned principally by the circumstance
that their navigation was mainly directed to the south and
south-west. On these southern voyages it had not escaped

TERRESTRIAL MAGNETISM. 131

them that the south end of the needle by which they
steered did not point precisely due south. We have even
the statement belonging to the 12th century, from one of
their determinations, of the amount of the " variation to the
south-east." (16°) The extension of the use of the mariners'
compass was much favoured by the very ancient connection
of China and India with Java,(161) and in a still greater
degree by the visits of people of Malay race to Madagascar,
and their settlements there.

Although, judging from the present very northerly position
of the magnetic equator, it is probable that when the mis-
sionary Guy Tachard remarked the existence of horary
variations of the declination in 1682 at the town of Louvo
in Siam, that place was nearly out of the northern magnetic
hemisphere, — yet it must be acknowledged that accurate
horary declination- observations were not made in the south-
ern magnetic hemisphere until a full century later. In 1794-
and 1795, John Macdonald followed the inarch of the
needle both at Tort Marlborough on the south-west coast
of Sumatra, and at St. Helena. (162) The results then
obtained called the attention of physicists to the great
decrease in the amount of the diurnal variation of the decli-
nation in the lower latitudes ; the differences between the
extremes at these stations amounting only to 3 or 4 minutes.
A more comprehensive and deeper knowledge of the phe-
nomenon was gained by the scientific expeditions of Preycinet
and Duperrey; but it is the establishment of magnetic ob-
servatories at three important points of the southern magnetic
hemisphere, — at Hobarton in Van Diemen Island, at St.
Helena, and at the Cape of Good Hope (at which places
hourly observations on the variation of the three elements

K 2

132 TERRESTRIAL MAGNETISM.

of terrestrial magnetism have been made on a uniform system
for several years), — which has first supplied general and com-
plete data. The march of the needle in the middle latitudes
of the southern magnetic hemisphere is quite the opposite
of its march in the northern hemisphere ; for as in the south,
the end of the needle which points towards the south moves
from morning to noon from the east towards the west, it
follows that the north-pointing end of the same needle is
moving at the same time from the west towards the east,
contrary to what we have described above as 'its march
during those hours in our own hemisphere.

Sabine, to whom we are indebted for the sagacious dis-
cussion of all these variations, has so combined five years
of hourly observation at Hobarton (42° 53' S. lat., 9° 57
E. decl.) and at Toronto (43° 39' N. lat., 1°33'W. decl.),
that the periods from October to February and from April
to August are distinguished from each other, the omitted
intervening months of March and September presenting, as
it were, transitional phaenomena. At Hobarton the north
end of the needle has daily two easterly and two westerly
maxima of elongation. (163) In the portion of the year
from October to February, the movement takes place towards
the east from 20h. or 21h. to 2h. ; there is then a small move-
ment towards the west from 2h. to llh., from llh. to 15h.
again to the east, and from 15h. to 20h. the needle returns
to the west. In the portion of the year from April to August
the eastern turning hours are retarded to 3h. and 16h., and
the western turning hours are made earlier, being at 22h.
and llh. In the northern hemisphere, the westward move-
ment of the needle from 20h. to Ih. is greater when the Sun
is in the northern than when he is in the southern signs :

TERRESTRIAL MAGNETISM. 133

in the southern hemisphere, where between those two hours
the direction of the movement of the needle is an opposite
one, its amount is greater when the Sun is in the southern
than when he is in the northern signs.

The question which I touched upon seven years ago in
Yol. i. of this work (164) , — i. e. whether there be a region of
the earth, perhaps between the geographical and magnetic
equators, in which, as a transitional region between the parts
of the earth in which the declination-needle moves at the
same hours in opposite directions, no horary variation might
be found ? — appears from subsequent experience, and es-
pecially from Sabine's perspicuous discussion of the obser-
vations at Singapore (1° 17' N. lat.), at St. "Helena
(15° 56' S. lat.), and at the Cape of Good Hope (33° 56'
S. lat.), to require to be answered in the negative. Hitherto
no place has been found at which the needle is without
diurnal movement (horary variation) ; and through the estab-
lishment of the magnetic stations we have been made aware
of the important ana very unexpected fact, that there are in
the southern magnetic hemisphere places at which the diurnal
variation of the magnetic needle appears to participate in the
phenomena, or to conform to the type, of either hemisphere
alternately. The Island of St. Helena is situated very near
to the line of weakest intensity of the earth's magnetic force,
in a part of the earth where that line recedes widely both
from the geographical equator and from the line of no
inclination. At St. Helena, the march of the north end of
the needle in the hours of the forenoon, in the months
from May to September, is the opposite of the march fol-
lowed by the same end of the needle in the corresponding
hours from October to February. According to five years

134 TERRESTRIAL MAGNETISM.

of hourly observation in the first-named portion of the year,
viz. from May to September, while the Sun is in the northern
signs, being the winter of the southern hemisphere, the
declination- needle points at 19h. furthest to the east of its
mean direction, and from that hour moves as it does in the
middle latitudes of Europe and North America towards the
west until 22h., and remains nearly stationary until 2h.
On the other hand, in the opposite portion of the year, from
October to February, the summer of the southern hemisphere,
when the Sun is in the southern signs, and theEarth is nearest
to the Sun, the needle points at 20h. most to the west, and
its movement from thence to noon is to the east, pre-
cisely in accordance with the type of Hobarton (in 42° 53'
S. lat.) and other places in the middle latitudes of the
southern hemisphere. At the time of the equinoxes, or soon
after, in March and April, and in the latter portion of Sep-
tember, the march of the needle fluctuates on different days,
forming periods of transition from the type of the southern
to tliat of the northern, and from that of the northern to
that of the southern hemisphere. (165)

Singapore is situated a little to the north of the geogra-
phical equator, between it and the magnetic equator, which
in this part of the earth is almost coincident, according to
Elliot, with the line of least force. According to the ob-
servations made every two hours at Singapore for the years
1841 and 1842, Sabine finds there the same opposite types
in the diurnal march of the needle from May to August, and
from November to February, as at St. Helena; and so also
at the Cape of Good Hope, although the latter station is 34°
from the geographical equator, and doubtless still further
from the magnetical one (the magnetic dip at the Cape

TERRESTRIAL MAGNETISM. 135

being 53° S.) ; and the Sun of course never passes through
its zenith. (166) We already possess several years of published
hourly observations, from which it appears that at this station,
almost precisely as at St. Helena, the needle from May to
September moves westward from its extreme eastern position
at 19^h. until 23-Ui.; and from October to March the
movement is, on the contrary, eastward from 204h. to l^h.
or 2h. The discovery of this phenomenon, of which the
existence is so well demonstrated, while at the same time
its causal connection is still veiled in such profound obscurity,
shows forcibly the importance of the system of hourly ob-
servations continued uninterruptedly for several years.
Disturbances which, as we shall presently see, deflect the
needle persistently, at one time to the west and at another
to the east, would prevent any well-assured inferences being
derived from isolated observations by travellers.

The extension of navigation, and the use of the compass
in geodesical surveys, led very early to the remark of oc-
casional extraordinary disturbances in the direction, accom-
panied by fluctuating, starting, or tremulous movements, of
the needles employed. This used to be ascribed to a certain
condition of the particular needle : it was very characteris-
tically termed, in the French nautical language, "Taffolement
de 1'aiguille," and it was recommended that " une aiguille
affolee" should be magnetised afresh, and more powerfully.
Halley was the first who stated the Aurora to be a magnetic
phenomenon (167), — on the occasion of his being asked by the
Royal Society of London, to explain the " great meteor" of
the 6th of March, 1716, seen over the whole of England, and
which, in his reply, he considered to be " analogous to that
to which Gassendi in 1621 had first given the name of Aurora

136 TERRESTRIAL MAGNETISM.

borealis." We know from his own statement that prior to
1716 Halley had never seen either a northern or a southern
Aurora ; although, in his voyages for determining the decli-
nation lines, he had advanced to the 52nd degree of south
latitude : — and southern Auroras are assuredly sometimes
seen within the tropics, as in Peru. It would appear, there-
fore, that Halley had never himself observed the remarkable
disturbances and fluctuations of the needle in conjunction
with seen (or unseen) northern or southern Auroras.
Olav Hiorter and Celsius, at Upsala, were the first who,
in 1741 (before Halley 's death), confirmed by a long
series of observations the connection between a visible
Aurora and disturbance from the normal march of the
magnetic needle, which was only conjectured by Halley.
This meritorious undertaking gave occasion to the first
concerted simultaneous observations in conjunction with
Graham in London : extraordinary disturbances of the de-
clination accompanying displays of Aurora were examined
specially by Wargentin, Canton, and Wilke.

Observations which I had the opportunity of making
together with Gay-Lussac in 1805, at Rome, on the Monte
Pincio, and more particularly a more extensive series which
I was thereby led to undertake in conjunction with Oltmanns,
at the equinoxes and solstices of 1806 and 1807, in a large
and well-detached garden at Berlin (employing a magnetic
telescope of Prony's, and a distant point of reference well illu-
minated by a lamp), — made me early aware that that part of
the telluric magnetic activity described under the general name
of " extraordinary disturbances," which acts so powerfully
at particular periods and in no merely local manner, deserved,
and from its complicated character required, a persistent

TERRESTRIAL MAGNETISM. 137

system of examination. The arrangement of the mark and
of the cross of wires in the telescope, which was suspended
by either a silken or metallic thread, and enclosed in a large
glass case, permitted readings to be taken to eight seconds
of arc. In this method of observation, the room in which
the magnetic telescope was placed might be left dark, thus
avoiding the disturbing effects of currents of air, which may
be occasioned by the means required for illuminating the
scale of otherwise excellent declinometers furnished with
microscopes. Entertaining the opinion which I then ex-
pressed, that " a continued uninterrupted hourly or half-
hourly process of observation (observatio perpetua) for several
days and nights was to be preferred to the detached observa-
tions of many months," we observed consecutively for five,
seven, or eleven days, and as many nights, (168) at the periods
of the solstices and equinoxes, epochs of which all the most
recent investigations have confirmed the great importance.
We soon recognised, that in order to study the proper phy-
sical character of these anomalous disturbances, it was not
sufficient to determine the measure or quantity of the devia-
tion from the change in the declination, but that there should
be added thereto a numerical estimate of the degree of in-
quietude of the needle, by noting the elongation of its os.
dilatory movements. "We found the ordinary diurnal march
of the needle so quiet, that among 1500 results drawn from
6000 observations, from the middle of May 1806 to the end
of June 1807, the oscillations were mostly only of half a scale
division, or 1' 12". The needle was often quite still,
or moving only 0'2 or 0'3 of a scale division during
very tempestuous and rainy weather; but on the arrival
of the magnetic storm, of which the more intense and later

133 TERRESTRIAL MAGNETISM.

manifestation is the Aurora or polar light, the oscillations were
from 14 to 38 minutes of arc, each oscillation being performed
in from 1 £ to 3 seconds of time. On many occasions the
magnitude and inequality of the oscillations, which extended
far beyond the limits of the divided scale on one or other, or
on both sides, made any observation impossible. (169) For ex-
ample, on the night of the 24th of September, 1806, this was
uninterruptedly the case from 14h. 40m. to 15h. 32m., and
again for the still longer interval from 15h. 57m. to 17h. 4m.
Most often in magnetic storms (the " unusual" or "larger
magnetic disturbances") the middle point of the arcs of
oscillation was in course of constant though unequal progress
to one or the other side, *. e. east or west ; but in other
and more rare cases, extraordinary oscillations were remarked
without any increase or decrease of the declination, — that is
to say, without displacement of the middle point of the
oscillations from its normal place at the particular hour.
After a long time of comparative repose, we sometimes saw
very unequal movements take place suddenly (describing
arcs of from 6 to 15 minutes, the different magnitudes being
either alternately or irregularly intermixed), and then the
needle would as suddenly become calm again. It was at
night that such alternations of total repose, and violent
fluctuation without progressive movement towards either
side, \\ere particularly striking. (17°) Another peculiar
modification of these movements should be mentioned ; it
occurred very rarely, and consisted in a particular kind of
affection of the inclination of the north end of the needle,
continuing for ] 5 or 20 minutes of time, with only very
moderate horizontal fluctuations, or even with an entire ab-
sence of any. Jn the registers of the British observatories,

TERRESTRIAL MAGNETISM. 139

where all the attendant circumstances are so diligently noted,
I find this purely vertical trembling ("constant vertical
motion, the needle oscillating vertically") mentioned three
times in the case of the declination magnet at Hobarton;(171)
and frequent notices of the vibration of the vertical and
horizontal force magnets at Toronto when no change took
place in their mean readings.

It appeared to me, that the average hour of the occurrence
. of the larger magnetic disturbances at Berlin was the third
hour after midnight, and that they ceased most usually at
5 A.M. We observed small disturbances in the afternoon
between 5 and 7 P.M., often on the same days in the
month of September when there ensued after midnight such
magnetic " storms/' that the magnitude and rapidity of the
oscillations made any reading, or any appreciation of the
middle point of the oscillation, impossible. I was very early
so convinced that the magnetic storms occur in groups, re-
turning on successive nights, that I announced this pecu-
liarity to the Berlin Academy, and invited friends to visit
me at predetermined hours, to enjoy the sight of the
phenomenon : and our. expectations were more often
realised than disappointed. (172) Kupffer, during his journey
to the Caucasus in 1829, and afterwards Kreil, in his valu-
able Prague observations, corroborated this disposition of
the magnetic storms to return at the same hours.^173)

Since the establishment of the British Colonial magnetic
observatories, the rich mass of materials which they
have afforded, and the able treatment of these materials
by Colonel Sabine, have caused the facts relating to the ex-
traordinary disturbances of the declination, which I had
recognised only in a general manner in my equinoctial and
solstitial observations of 1806, to become one of the most

140 TEERESTRIAL MAGNETISM.

important of the acquisitions which have been gained towards
a complete knowledge of terrestrial magnetism. Sabine
has distinguished the disturbances, in the results of both
hemispheres, into classes, according to the hours of their
occurrence, whether in the day or in the night, — the seasons
of the year, — and the direction in which they deflect the
needle, whether to the east or to the west. At Toronto
and at Hobarton, the disturbances were twice as great, both
in frequency and value, in the night as in the day,(174) as
in the older observations at Berlin; but in from 2600 to
3000 disturbances at the Cape of Good Hope, and more
particularly at St. Helena, the direct contrary was the case,
as shown by a similarly thorough investigation of the phse-
nomena at those observatories by Captain Younghusband.
At Toronto, the principal disturbances were observed, on the
average, from midnight to 5 in the morning ; occasionally
only they were observed earlier, between 10 P.M. and mid-
night,— predominating, therefore, in the night at Toronto as
at Hobarton. After a very laborious and well-devised examina-
tion of 3940 disturbances (of the declination) at Toronto, and
3470 at Hobarton, taken from six years of observation, 1843
— 1848, (the observations selected as disturbed making the
ninth and tenth portions of the entire mass), Sabine has
drawn the conclusion, (175) that the disturbances belong to a
peculiar class of periodically recurring variations, following
recognisable laws dependent on the position of the Sun in
the ecliptic, and on the diurnal rotation of the Earth upon its
axis ; that they ought no longer to be called irregular
movements ; and that we may distinguish in them, together
with a particular local type, general processes affecting the
whole globe. In regard to the different amount of disturbance
in different years, the same years in which the disturbances

TERUESTUIAL MAGNETISM. 14-1

were most frequent at Toronto in the northern hemisphere
were also those in which they were most numerous at
Hobarton in the southern hemisphere, — and that nearly in
an equal degree. In regard to the differences in different parts
of the year, the disturbances were, on the whole, twice as
numerous at Toronto in the summer months of the station
(April to September), as in its winter months (October to
March). (In my Berlin observations of 1806 the greatest
number in any month belonged to the month of September
at the time of the autumnal equinox.) (176) They are more
rare in the winter months of each station ; i. e, more rare from
November to February at Toronto, and from May to August
at Hobarton. The times of the Sun's passage over the equator
are also, according to Captain Younghusband, in a high degree
remarkable for the greater frequency of disturbances at St.
Helena and the Cape of Good Hope. In the higher latitudes
of both hemispheres the disturbances are less frequent in the
winter than in the summer, i.e. less frequent at Hobarton from
May to September, and at Toronto from November to March.
A most important feature in the phenomenon also first
made known by Sabine, is the regularity, in both hemi-
spheres, of the disturbances in the direction of the needle
either to the east or to the west. At Toronto, where there
was a small west declination (1° 33'), the number of distur-
bances which deflected the needle to the east preponderated
in summer (June to September), and that of those which
deflected it to the west, in winter (December to April) ;
and this in no inconsiderable ratio — 411 : 290. The case
is similar at Hobarton, according to the seasons of the
southern hemisphere. In the winter months of that station
(May to August) the disturbances are strikingly less fre-

142 TERRESTRIAL MAGNETISM.

quent.(177) The analysis of six years of observation at the two
opposite stations of Toronto and Hobarton, has conducted
Colonel Sabine to the remarkable result, — that in both
hemispheres, not only the number of disturbances, but also
the amount of average diurnal variation from the mean
or normal place and the latter (obtained from all the obser-
vations of the year omitting the 3469 storms), increased gra-
dually in the five years from 1843 to 1848, from 7''65 to
10'-58; and that simultaneously with this .a similar increase
could be traced in the variations of the inclination and of the
total force. This result gained a yet higher degree of impor-
tance when he found both a confirmation and a generalisation
of it in Lament's detailed investigation of Sept. ] 851, "on a
decennial period presented by the diurnal movement of the
declination needle." According to observations at Gottingen,
Munich, and Kremsmiinster, (178) the mean amount of the
diurnal variation of the declination was at a minimum from
1843 to 1844, and at a maximum from 1848 to 1849.
After the declination has increased five years, it decreases
for the same number of years, as is shown by a series of
observations which lead back to a maximum in 17864-.(1?y)
In order to find a general cause for such a periodicity
in all the three elements of terrestrial magnetism, one is
inclined to have recourse to some cosmical connection.
Such a connection is found, according to Sabine's con-
jecture,^80) in the variations which take place in the photo-
sphere of the Sun, i. e. in the luminous gaseous envelopes
of the dark solar orb ; for, according to Schwabe's researches,
extending over a long series of years, the periods of greatest
and least frequency of the solar spots coincide perfectly with
those discovered in the variations of terrestrial magnetism.

TERRESTKlAL MAGNETISM. 143

Sabine was the first to call attention to this accordance or
coincidence, in his paper presented to the Royal Society of
London in March 1852. Schwabe said, in a passage with
which he enriched the astronomical portion of my Cosmos,
f: There is no doubt that, at least from the year 1826 to
1850, the solar spots have shown a period of about ten years
with maxima in 1828, 1837, and 1848, and minima in
1833 and 1843." (181) Sabine also sagaciously adduces, in
confirmation of the powerful direct influence of the Sun upon
terrestrial magnetism, the remark that in both hemispheres
the period when the intensity of the Earth's magnetic force is
greatest, and the direction of the needle most near to the ver-
tical, is in the months from October to February, which is just
the period when the Earth is nearest to the Sun, and is moving
with the greatest velocity in its orbit. (182)

The simultaneity of many magnetic storms, apparently
propagated instantaneously between places many thousand
miles apart, and even almost round the entire globe (as on
the 25th of September, 1841, between Canada, Bohemia,
Cape of Good Hope, Van Diemen Island, and Macao), has
been already treated of in Vol. i.(183) : examples were also
cited of cases in which the perturbation appeared to have a
more local character, extending from Sicily to Upsala, but not
beyond Upsala to Altyn or Lapland. In the simultaneous
observations of declination arranged by Arago and myself in
1829, to be made with similar instruments of Gambey's at
Berlin, Paris, Freiberg, St. Petersburg, Kasan, and N icolaieff',
some considerable perturbations showed themselves at Berlin
which did not extend to Paris, or even to a mine at Freiberg
where Reich was making subterranean magnetic observations.
Great deflections and oscillations of the needle accompanying

144 TERRESTRIAL MAGNETISM.

Auroras at Toronto, were synchronous with magnetic
storms at Kerguelen Island in the southern hemisphere,
but not at Hobarton. Considering the all-pervading
character of magnetic as well as gravitating force in all
matter, it seems indeed difficult to form a clear concep-
tion of any obstacles to the propagation of magnetic
force through the globe, analogous to those which im-
pede the undulations of sound, or those which intercept
earthquake waves, so that at some nearly adjacent places
the earthquake shocks have never been known to occur
simultaneously. (184) May it be that the intersection of
magnetic impulses may oppose their propagation ?

I have described both the regular and the seemingly
irregular movements presented by horizontally-sus-
pended needles. Having examined the regular periodi-
cally-recurring march of the needle, and found the mean
direction around, or, in other words, on either side of
which the oscillations have taken place — say from one
solstice to the return of the same solstice in the follow-
ing year — this mean direction is the magnetic meridian
of the place for the given year. The comparison of the
angle made by the magnetic meridians of different
places on the globe with their geographical meridians,
first led to the recognition of " variation-" (i. e. declina-
tion-) lines of strikingly different values (which Andrea
Bianco in 1436, and Alonso de Santa Cruz, the cosmo-
grapher of the Emperor Charles V., thus early attempted
to enter on charts); and subsequently to the happily
devised generalisation of isogonic curves, or lines of equal
declination, to which the grateful recollection of English
navigators long gave the historic name of "Halleyan
lines." Among these lines of various curvature, running

TERRESTRIAL MAGNETISM. 145

in some places almost parallel to each other, more rarely
returning into themselves, and, where they do so, form-
ing closed systems of an oval form, especial attention has
been given to the lines on which the declination is 0°,
at all points of which the direction of the compass-
needle is that of true or geographical north and south,
— and in receding from which lines on either side,
opposite declinations are found (i. e. east declinations
on one side, and west on the other), the amount of
declination increasing unequally with increased dis-
tance. (185) I have shown elsewhere how Columbus 's
first discovery of a "line without variation" in the
Atlantic Ocean, on the 13th of September 1492, gave an
impulse to the study of terrestrial magnetism, - which,
however, for two centuries and a half was directed
solely towards the improvement of ships' reckonings.

Greatly as in modern times the higher scientific
training of navigators, and the improvements which
have taken place in instruments and in methods, have
extended our knowledge of detached parts of the lines
of no declination in the north of Asia, the Indian
Archipelago, and the Atlantic Ocean, yet in this de-
partment of knowledge, where the want of a general
or cosmical view is felt, there is still reason to com-
plain of slowness of progress, and absence of the desired
completeness in the research. I am aware that a count-
less number of observations, made during accidental
passages across the lines of no declination, have been
entered into ships' journals ; but comparison and com-
bination of the materials are wanting ; — nor can the
results of such a labour (even supposing it to be under-
taken), either on this subject, or on that of the position

VOL. IV. L

146 TERRESTRIAL MAGNETISM.

of the line of no inclination for a definite epoch, have
all the desired importance, until ships charged with
the sole duty of tracing these lines uninterruptedly
throughout their course shall be sent contempora-
neously to different seas. I here renew a solicitation
to which I have repeatedly permitted myself to give
free expression. (186) For systematic investigations in
terrestrial magnetism it is of primary importance that
the determinations should be contemporaneous.

According to our present general knowledge of the
" lines of no declination," instead of four such lines
running as meridians from pole to pole, as was sup-
posed at the end of the 16th century, (187) there appear
to be three very differently shaped systems, using the
word " system " in this case to denote a group of
isogonic lines which include a line of no declination
unconnected, so far as we know, with any other such
line. Of these three systems, which I shall presently
describe separately, the Atlantic one consists simply of
a line of no decimation directed from SSE. to NNW.,
and recognised as extending from the 65th degree of
south to the 67th degree of north latitude. The second
system is situated fully 150 degrees to the eastward of
the first (looking in both cases only at the points of
intersection of the lines of no declination with the
geographical equator) ; it is the broadest and most
complex system, and occupies the whole of Asia and
Australia. It has one summit directed to the north,
and one to the south: in the north-eastern part the
line of no declination takes an oval shape, and returns
into itself, surrounding lines of successively and rapidly
increasing declination. The east and west sides of this

TERRESTRIAL MAGNETISM. 147

ellipse, like the Atlantic of 0°, have a direction from
south to north, and between the Caspian and Lapland,
from SSE. to NNW. The third system, that of the
Pacific Ocean, which has been least examined, is less
than the others, and is almost entirely confined to the
south of the geographic equator : it forms a closed oval
of concentric lines of successively decreasing east de-
clination. On the African continent, inferring from
what we know of the declination on its coasts, (188)
there are only lines having a west declination of from
6° to 2 9°; the Atlantic line of no declination quitted
the south point of Africa (Cape of Grood Hope), ac-
cording to Purchas, as early as 1605, moving from east
to west. As far as we are acquainted with the facts,
there is not in the interior of central Africa an oval
group of concentric declination lines similar to that in
the Pacific.

The Atlantic part of the American curve of no de-
clination has been accurately determined and laid down
in an excellent work by Colonel Sabine, "On the De-
clination Lines over the Atlantic Ocean, between the
parallels of 60° north and 60° south latitude, for the
year 1840," based on 1480 observations either made in
that year, or reduced to it by careful allowance for
secular change. Tracing this line of no declination
from its most southern known point (in lat. 70° S.,
where it was found in about 19° W. long.), (189) it runs
NNW. to 3° west of Cook's Sandwich Land, and to 9|°
east of South Georgia ; it then approaches the Brazilian
coast, which it enters at Cape Frio, two degrees east of
Rio Janeiro ; cootinues within the South American
continent to S. lat. 0° 36', and quitting the continent

L 2

148 TERRESTRIAL MAGNETISM.

again a little to the east of Para, near Cape Tigioca, at
one of the mouths of the Amazon (Rio de Para), it cuts
the geographic equator in 48° 30' W. long., and runs at
a distance of nearly ninety miles from the coast of
GKiyana up to 5° of North latitude, from whence, follow-
ing the curve of the smaller West-Indian Islands, it
ascends to 18° N. lat., and passes from thence to the coast
of North Carolina, which it enters in lat. 34° 50'9 W. long.
76°*30/, near Cape Lookout, south-west of Cape Hatteras.
In the interior of North America this curve continues its
north-west direction to lat. 41-1-0, long. 80°, towards
Pittsburgh, Meadville, and Lake Erie. There is reason
to suppose that it may already have moved, in some
parts of its course, more than a degree to the west since
1840.

The Australo- Asiatic curve of no declination, — if, with
Erman, we regard the line which at Kasan trends north-
wards to Archangel and Eussian Lapland, as part of the
same line which passes the Moluccas and the Sea of
Japan, — can scarcely be traced in the southern hemi-
sphere so far as the 62d degree of latitude. Its most
southern known point is more to the west of Van
Diemen Island than was previously supposed. The
three points at which Sir James Ross, (19°) in his ant-
arctic voyages of discovery, crossed it in 1840 and 1841,
in the parallels of 62°, 54^-°, and 46°, were all between
the longitudes of 133°-20' and 1350-4(X East, showing an
almost north and south direction in this part of its
course. It then traverses Western Australia, entering
the south coast of Nuyts Land (about ten degrees west
of Adelaide), and quitting the northern coast near
Vansittart River and Mount Cockburn. From hence it

TERRESTRIAL MAGNETISM. 149

enters the Indian Archipelago, where the declination,
inclination, total force, and line of minimum intensity of
the total force, and of maximum intensity of the hori-
zontal force, have all been investigated with admirable
exactness, in 1846 — 1848, by Captain Elliot. Here the
line of no declination passes immediately to the south of
the Island of Flores, through the small " Sandal Wood "
Island, (191) and from 120° to 94° E. long, in a due east
and west direction; — as, indeed, Barlow had laid it down
correctly sixteen years before. From the last-named
meridian, and about 9^° S. lat., it seems probable —
judging by the line of 1° E. declination which Captain
Elliot has traced up to Madras — that the line of no
declination takes a new direction trending to the north-
west. It is impossible at present to form any certain
conclusion as to whether, in its further prolongation, it
cuts the geographic equator in the meridian of Ceylon,
and enters the continent of Asia either between the Ghilf
of Cambay and Gruzerat, or further to the west in the
Bay of Muscat, (l92) and so is identical with the curve
of no declination, (193) which appears to run to the south-
ward from the shores of the Caspian ; — or whether, on
the contrary (as Erman thinks), it bends back to the
east, and passes between Borneo and Malacca, and,
ascending towards the north, reaches the Sea of
Japan, (l94) and enters North-eastern Asia through the
Sea of Ochotsk. It is greatly to be regretted that the
very numerous determinations which must have been
obtained in the frequent navigation to India, Australia,
the Philippines, and the north-east coast of Asia, have
not yet been co-ordinated and rendered conducive to
general views, by connecting the south of Asia with its

150 TERHESTKIAL MAGNETISM.

better magnetically explored northern regions, or to the
solution of questions which have been raised since
1840. Not to mix the uncertain with the certain, I will
therefore limit myself in the following description to the
Siberian portion of the Asiatic continent, so far as it is
known to us down to the parallel of 45° by the observa-
tions of Erman, Hansteen, Due, Kupffer, Fuss, and
myself. There is no other part of the earth's surface on
which magnetic lines could have been traced over so
great an extent of continental territory, — a circumstance
which appears to me of some importance, as it did long
since to Leibnitz when speaking of the importance of
the Russian dominions in Europe and Asia in reference
to this subject. (195)

Advancing from the west towards the east as the
usual direction followed by European expeditions to
Siberia, and commencing at the northern part of the
Caspian Sea, we find that, at what may be termed a
group of stations, viz. the little island of Birutschikassa at
Astrachan, the Elton Lake in the Kirghis steppe, and
Uralsk on the Jaik, — places situated between the
latitudes of 45° 43' and 51° 12', and the East longitudes
of 46° 37' and 51° 24', — the declination ranges from
0° 10' E. to 0° 37' W. (196) Further to the north this
line of declination inclines rather more to the north-
west, passing near Nishnei Novgorod, (197) — (in 1828,
between Osablikowo and Doskino, in lat. 56°, long. 43°).
It is prolonged towards Russian Lapland, passing
between Archangel and Kola, or more exactly, accord-
ing to Hansteen in 1830, between Umba and Ponoi. (198)
In advancing eastward from this point, between the
parallels of 50° and 60°, it is not until we have traversed

TERRESTEIAL MAGNETISM. 151

nearly two thirds of the breadth of Northern Asia in this
its widest part (a space now entirely occupied by east
declination), that we arrive at a line of no declination,
which, at the north-eastern part of Lake Baikal, west of
Viluisk, ascends towards a point situated in the meridian
of lakutsk, or in about 130° E. long., and as much as
68° N. lat., from whence it redescends towards Ochotsk
(in 143° 10' E. long.), forming the outermost line of the
eastern side of the oval-shaped group of concentric
declination-lines which has been before alluded to, and
then crosses the Kurile Islands, passing south into the
Sea of Japan. The isogonic lines from 5° to 15° east
declination, which occupy the space between the east
and west Asiatic lines of no declination, are all concave
towards the north. Their greatest curvature falls,
according to Erman, in long. 80°, or in a meridian
nearly intermediate between Omsk and Tomsk, and not
very different from that of the southern point of the
peninsula of Hindostan. The longer axis of the closed
oval group extends through twenty-eight degrees of
latitude, or about to the Corea.

A similarly shaped group or system of declination-
lines, but of larger dimensions, presents itself to our
observation in the Pacific. The closed curves there
form an oval extending from 20° N. to 42° S. latitude.
The principal axis is situated in 130° W. long. The
most remarkable feature of distinction between this
singular group (the greater portion of which is in the
southern hemisphere, and which is wholly oceanic), and
the previously described continental group of Eastern
Asia, is that in the Pacific group it is East declination
which decreases, while in the Asiatic oval it is West

152 TERRESTRIAL MAGNETISM.

declination which increases, in penetrating towards the
interior of the oval. In the Pacific oval we only know
at present declinations of from — 8° to — 5°. May there
be, more in the interior, another line of no declination
enclosing a space of westerly declination ?

The curves of no declination, like all the magnetic
lines, have each their own proper history, although . it
can only in any case be very imperfectly traced back
beyond two centuries. We possess, indeed, detached
materials of an earlier date, going back even to the
14th and 15th centuries. Hansteen has here also the
great merit of having assembled and compared the data
with great judgment. It would appear as if the north
magnetic pole were moving from west to east, and the
south magnetic pole from east to west : but we know
from exact observations, that different portions of the
isogonic curves advance at very unequal rates of pro-
gression,— that where they were once parallel, they are
now no longer soy — and that adjoining regions, over
which decimations of opposite denominations prevail,
enlarge and contract in very different directions. The
lines of no declination in Western Asia and the Atlantic
Ocean advance from east to west. The former (or West
Asiatic) passed through Tobolsk in about 1716 ; Catha-
rinenburg (in Chappe's time) in 1761 ; and between
Osablikowo and Doskino (not far from Nishnei Nov-
gorod) in 1829; so that in 113 years it had advanced
24 f° towards the west. If the line of no declination of
the Azores, determined by Columbus on the 13th of
September 1492, was the same line which, according to
the observations of Davis and Keeling, passed in 1607
through the Cape of Grood Hope, (199) and the same

POLAR LIGHT, OR AURORA. 153

which we now see in the western part of the Atlantic
Ocean, passing from the mouth of the Amazons to the
coast of North Carolina ; — we are led to ask what has
become of the line of no declination which passed in
1600 through Konigsberg; in 1620? through Copen-
hagen; in 1657 — 1662 through London (and yet, ac-
cording to Bicard, only in 1666 through Paris, which is
to the east of London) ; and rather before 1668 through
Lisbon ? (20°) There are particular parts of the globe
in which, throughout a long interval of time, no secular
change has been remarked. Sir John Herschel has
called attention (201) to such a long period of suspension
of change in Jamaica, as Euler (202) and Barlow (203) had
already done to a similar one in South Australia.

Polar Light, or Aurora.

I have treated in detail the three elements of Ter-
restrial Magnetism, or the magnetic "Declination," "In-
clination," and "Force," in their connection with
geographical position, and their variations according to
hour and season. The " extraordinary disturbances," or
perturbations, which were first observed in the decli-
nation, are, as Halley conjectured, and as Dufay and
Hiorter recognised, partly the precursors and partly the
accompaniments of the magnetic polar light, or aurora
borealis or australis. In the "Eepresentation of Nature"
contained in the first volume of my work, I have dis-
cussed with some degree of fullness the peculiarities
belonging to this " telluric luminous process," often so
remarkable by its beautiful display of colours; and

154 TERRESTRIAL MAGNETISM.

more recent observations have in general been favour-
able to the views there expressed, in which " the aurora
was not regarded as the cause of the disturbance in the
equilibrium of the distribution of the earth's magnetism,
but rather as the result of a state of telluric activity
excited to the point of the production of a luminous
phenomenon ; an activity manifesting itself on the one
hand by the fluctuations of the needle, and on the other
by the appearance of the brilliant auroral light." In,
this view the polar light might appear as a kind of
silent discharge, — as the termination of a magnetic
storm. In electric storms the equilibrium of the dis-
turbed electricity is also restored by a development of
luminosity, — by the flashing lightning, accompanied by
the noise of thunder. In a phenomenon so complex and
mysterious, the repeated bringing forward of a deter-
minate hypothesis (204) has at least the advantage of
inviting a more sustained and careful observation of all
those details by which it may be confuted or corroborated.
In dwelling on the purely objective description of
these phenomena, and availing myself principally of
the fine uninterrupted eight-months series of examina-
tions for which we are indebted to the sojourn of several
distinguished physicists (205) in the extreme north of
Scandinavia in 1838 — 1839, I would direct the atten-
tion in the first place to the dark misty wall which rises
gradually on the horizon, and is called the "black
segment of the aurora borealis." (206) The blackness, as
Argelander remarks, is not a result of contrast, for it is
sometimes seen before the luminous arch begins to
bound it: — it is a process going forward in apart of the
atmosphere, for hitherto we have had no evidence of

POLAR LIGHT, OR AURORA. 155

any admixture of foreign matter causing the darkness.
The smallest stars are perceived by the telescope in the
dark segment as well as in the coloured and bright parts
of the fully developed aurora. The black segment
seems to be much more rare in the higher than in the
middle latitudes : the observers above spoken of did not
see it once during the whole of February and March,
although auroras were abundant and the sky extremely
serene ; nor did Keilhau see it once during an entire
winter passed by him at Talving in Lapland. By exact
determinations of stars' altitudes, Argelander has shown
that they are not in the least influenced by any part of
the aurora. Even out of the segment there appear,
although rarely, black rays, which Hansteen (207) and I
have more than once seen stream upwards: together
with these, there appear roundish black patches, enclosed
by luminous borders, to which Siljestrom has given
especial attention. (208) Also, in that rare phenomenon,
the auroral corona, which, by the effect of linear
perspective projection, corresponds in altitude to the
magnetic inclination of the place, the middle of the
corona is mostly a very deep black. Bravais regards both
this and the black rays merely as optical effects of con-
trast. Several luminous arches often appear at the same
time — in rare cases as many as seven or eight, advan-
cing parallel to each other towards the zenith; sometimes
they are entirely absent. The bundles of rays and lumi-
nous streamers assume the most varied forms : curved,
wreathed like garlands, indented, hooked, in waving
sheets, or like ships' ensigns floating in the breeze. (209)
In high latitudes, " the usually prevailing colour of
the polar light is white, milk-white indeed when the

156 TERRESTRIAL MAGNETISM.

intensity is feeble. As the tone of colour becomes more
vivid it passes into yellow, the middle of the broad beam
becoming full yellow, and detached red and green
appearing at either margin. When the beams are long
and narrow, the red is above and the green below.
When the motion is lateral from left to right, or vice
versa, the red always appears on the side towards which
the movement is directed, and the green remains behind."
Of these complementary colours, green and red, it is
very rare to see one without the other. Blue is not seen
at all ; and a dark red, resembling the reflection of a
fire, is so rare in the North, that Siljestrom only saw it
once.(210) The illuminating power of the aurora borealis
never quite equals, even in Finmarken, that of the full
moon.

The connection so long maintained by me as probable,
of the polar light with the formation of the " smallest
and most attenuated cirrus, or light fleecy clouds, the
parallel equidistant lines of which most often follow
the direction of the magnetic meridian," has indeed
found in more recent times many defenders ; but whether,
as the northern traveller Thienemann and Admiral
Wrangel think, these lines of cirri form the substratum
of the aurora, or whether they are not rather, as has
been the opinion of Franklin, Kichardson, and myself,
the effects of a meteorological process accompanying and
produced by the magnetic storm, still remains unde-
cided. (211)

Besides the general conformity of the direction of the
magnetic declination with that of the regularly arranged
and very delicate cirrus ("bandes polaires"), my atten-
tion was strongly attracted, both in Mexico in 1803

POLAR LIGHT, OR AURORA. 157

in Asia in 1829, by the rotation, or movement in
azimuth, of the points of convergence. When the
phenomenon is very complete, the two apparent points
of convergence do not remain fixed, the one in the
north-east, the other in the south-west, (in the direction
of the line which unites the highest points of the
nocturnally bright auroral arches,) but the line turns, so
as gradually to bring those points more towards the east
and west.(212) A perfectly similar movement in the line
which joins the summits of true auroral luminous arches
(the ends of feet changing in azimuths from E. — W.
towards IS". — S.) has been observed with great exactness
more than once in Finmarken. (213) In the view here
taken, the cirri arranged as "polar bands " correspond in
position to the " streamers " or " bundles of rays " which
shoot upwards towards the zenith from the auroral arch
of which the span is most often E. — W., and are not
therefore to be confounded with the arch itself; the
latter was once seen by Parry in the arctic regions
during daylight, it having continued visible after a
night of aurora. A similar phenomenon was witnessed
on the 9th of September 1827 in England. It was
even possible to recognise the streamers ascending from
the arch in daylight.(-14)

It has been often stated that a perpetual evolution of
light, or aurora, takes place around the north magnetic
pole. Bravais, who observed without intermission for
200 nights, in which 152 auroras admitted of exact
description, does indeed assure us that nights without
any aurora are very exceptional ; but yet he sometimes
either saw no trace of auroral light during a very clear
night, and with an uninterrupted horizon, or the " mag-

158 TERRESTRIAL MAGNETISM.

netic storm " only began very late. The greatest abso-
lute number of auroras were seen towards the latter part
of the month of September, and as March also showed
relative frequency as compared with February and
April, we may surmise here also, as in other magnetic
phenomena, a connection with the equinoxes. In addi-
tion to instances of aurora borealis seen in the southern
hemisphere in Peru, and of aurora australis see,n so far
north as Scotland, I may mention a coloured aurora
borealis' seen and observed for two hours by Captain
Lafond in the ' Candide' on the 14th of January 1831,
south of New Holland, in latitude 45°. (215)

The sound sometimes attributed to the aurora has
been negatived by the French physicists, and by Siljes-
trom at Bossekop, (216) as decidedly as by Thienemann,
Parry, Franklin, Richardson, Wrangel, and Anjou.
Bravais estimated the elevation or height of the aurora
above the earth at 100,000 metres, or more than fifty
miles; while a very meritorious observer, Mr. Farqu-
harson, had estimated it as only four thousand feet.
The grounds on which all these estimations are based
are exceedingly uncertain, and are liable to be altogether
vitiated as well as by optical illusions, as by improved
assumptions of the actual identity of a luminous arch
seen simultaneously from two distant places. On the
other hand, the influence of the aurora on the declina-
nation, inclination, and intensity, both horizontal and
total, of the magnetic force (on all the elements, there-
fore, of the earth's magnetism), is undoubted, although
very unequal at different stages of the phenomenon,
and in the effect produced on the different elements.
The most complete observations on this subject are

POLAR LIGHT, Oil AURORA. 159

those of Siljestrom(217) and Bravais (1838—1839) in
Lapland, and those at Toronto in Canada (1840 — 1841),
discussed with much perspicacity by Sabine. (218) In
our concerted simultaneous observations made at Berlin
in Mendelssohn's and Bartholdy's garden, at Freiberg
below the earth's surface, at Petersburg, Kasan, and
Nicolaieff, a derangement of the magnetic declination
was perceived at all the stations during an aurora seen
at Alford in Aberdeenshire (lat. 57° 15') on the 19th
and 20th of September 1829; — and at those among
them at which the other magnetic elements were also
observed, the dip and force were also seen to be
affected. (219) During the fine aurora borealis observed
at Edinburgh by Professor Forbes, on the 21st of
March 1833, the magnetic inclination in the mine at
Freiberg underwent a remarkable decrease, and the
declination was so disturbed that it was hardly possible
to take any readings. A phenomenon which appears
deserving of particular attention, is a decrease of the
total magnetic force during increasing activity of the
auroral process. The results obtained by Oltmanns
and myself at Berlin, during a fine aurora borealis on the
20th of December 1806, (22°) and which were printed
in Hansteen's " Untersuchungen iiber den Magnetismus
der Erde," have been corroborated by Sabine and by the
French physicists in Lapland in 1838.(221)

As in the foregoing careful attempt at an account of
the state of our positive knowledge in regard to the
phenomena of terrestrial magnetism, I have limited
myself to a simply objective representation, inasmuch
as theoretical views drawn by inductive reasoning from
analogies have not yet assumed a sufficiently satisfactory

160 TERRESTRIAL MAGNETISM.

form, — I have no less designedly avoided the attempt to
infer geognostical connections between the direction of
great mountain-chains or stratified rocks, and of the
magnetic lines, more especially the isoclinal and isody-
namic lines. I am far from denying the possible
influence of cosmical primeval forces, of dynamic and
chemical forces as well as of magnetic and electric
currents, on the formation of crystalline rocks and the
filling up of veins ;(222) but seeing the progressive
movement, accompanied by change of form, of all the
magnetic lines, it cannot be supposed that the position
which they may be found to occupy at any particular
time, can throw any light on the relations of direction
of mountain-chains upheaved at exceedingly remote
but very different epochs, or on the foldings or corruga-
tions then taking place in the gradually hardening,
heat-exhaling crust of the earth.

There are indeed other relations, not affecting the
earth's general magnetism, but only of a very partial
and local character, subsisting between geognostic and
magnetic phenomena, which may be called " mountain-
or rock-magnetism." (223) I was much interested by
these when examining in 1796 (previous to my depar-
ture for America) the magnetic serpentine rock of the
Haidberg in Franconia, and they then gave rise in
Germany to a good deal of literary debate. They
present a series of problems very accessible to observa-
tion and experiment, in which there remains much to
be determined. The intensity of the rock-magnetism
may be tested in detached fragments of hornblende- and
chloride-slate, serpentine, syenite, dolerite, basalt, mela-
phyre, and trachyte, by experiments of deflection aiid

POLAR LIGHT, OR AUllOKA. 161

vibration of magnetic needles. In this manner, by
comparison of the specific weight, and by applying the
microscope to finely pulverised portions, it may be
decided whether the strength of the polarity does not
often proceed less from the quantity of the interspersed
grains of magnetic iron and iron oxide, than from the
relative position of the grains. A more important
question, however, in cosmical respects, and which was
propounded by me long since in connection with the
Haidberg, is: whether there are mountain-ridges in
which an opposite polarity is found on the opposite
sides ? (224) An accurate astronomical determination of
the magnetic axis of such a mountain would become of
great interest, if after considerable intervals of time
there should be recognised either an alteration in the
direction of the axis, or an (at least apparent) indepen-
dence of the three variable elements of the general
magnetic force of the earth.

VOL. IT. M

162 KEACTION OF THE INTERIOR OF THE EARTH

II.

Reaction of the interior of the earth on its exterior, showing itself,

— (a) dynamically, by waves of agitation, or " earthquakes ; "

— (Z>) by the increased temperature of the water of springs,
and the admixture of other substances, as salts and gases, in
mineral and thermal springs ; — (c) by the breaking forth of
elastic fluids, accompanied at times by phenomena of spon-
taneous ignition (gas- and mud-volcanoes, burning naphtha,
and salses) ; — (d) by the grand and powerful action of
volcanoes proper, in which, (by permanent channels of com-
munication with the atmosphere through fissures and craters),
earths molten at profound terrestrial depths are erupted as
glowing scoriae, partly subjected, at the same time, to pro-
cesses whereby they are changed into crystalline rocks, and
partly poured forth in long narrow streams of lava.

IN order to maintain, in accordance with the fundamental
plan of my work, the connecting links uniting all telluric
phenomena, and to preserve the representation of the
concurrent action of forces in a single system, we must
here recall how, beginning from the general properties
of matter, and the three principal directions of their
activity (attraction, light- and heat-exciting undulations,
and electro-magnetic processes), the first division of the
present volume treated of the magnitude, figure, and
density of our planet ; its internal distribution of heat ;
and its magnetic condition. The above-named classes
of activity in the material universe (molecular and
mass-attraction or gravitation, light- and heat-exciting

ON ITS EXTERIOR. GENERAL REMARKS. 163

undulations, and electro-magnetic activity) are nearly
allied manifestations (225) of one and the same primary
force, among which the first is the most independent of
diversity of substance. We have in all the above
respects represented our planet in its cosmical relation
to the central body of its system : its primitive internal
heat, generated possibly by the condensation of a rotating
nebulous ring, is modified by the solar action ; and, as
has been above stated, according to the latest hypotheses
a periodical action is exercised on terrestrial magnetism
by a particular condition of the Sun, manifested to us
by the periodically varying frequency of the solar spots,
or openings in the outer envelopes of the Sun.

This second division of the present volume will be
devoted to the group of telluric phenomena which are
attributed to the reaction, still going forward, of the in-
terior of the earth on its exterior.(226) I designate the
whole of these phenomena by the general name of volca-
nism or volcanicity, and I regard it as an advantage not
to divide effects having the same causal connection, and
differing only by the strength of the manifestation of
the acting force, and by varieties in the complication of
the physical processes involved. In this generality of
view, small and apparently insignificant phsenomena
acquire a greater significance. An observer not scienti-
fically prepared, who visits for the first time a basin
filled by a hot spring, and sees ascend from it gases
which extinguish the flame of a candle, or walks
between rows of variable cones of mud-volcanoes
hardly exceeding in height la is own stature-, would
not divine that the place now thus harmlessly occupied
has been repeatedly the scene of fiery eruptions as-

M 2

164 KEACTION OF THE INTERIOR OF THE EARTH

cending to the height of many thousand feet, and that
the same internal force is at work as that which gives
rise to colossal craters of elevation, and even to the
mighty devastating, lava-pouring volcanoes of Etna and
the Peak of Teneriffe, and to those of Cotopaxi and
Tunguragua from which scoriae are ejected.

Among the manifold and cumulative phenomena of
the reaction of the interior against the external crust,
I will first consider separately those whose essential
character is simply dynamical, and which consist of un-
dulations of the solid strata, — being a volcanic activity
not necessarily accompanied by chemical alteration of
substances, or by the eruption or production of any fresh
substance. In the other phenomena of the reaction of
the interior against the exterior, — in gas- and mud-
volcanoes, in burning naphthas and salses, and in the
great burning mountains which were first, and for a
long time exclusively, termed volcanoes, — there are
always present, the production of some fresh sub-
stance (gaseous or solid), processes of decomposition
and evolution of gases, and the formation of rock from
particles arranged in crystalline order. These are,
speaking in the most general terms, the distinctive
marks or signs of the volcanic vital activity of our
planet. In so far as this activity is to be principally
ascribed to the high temperature of the interior of the
earth, it is probable that all heavenly bodies which have
been condensed from a gaseous to a solid state, with an
accompanying enormous extrication of heat, must
present analogous phsenomena. The little that we
know of the configuration of the Moon's surface seems
to afford indications (2-7) in correspondence with this

ON ITS EXTERIOR. GENERAL REMARKS. 165

view. We can imagine upheavings and formative acti-
vity in the production of crystalline rocks from a molten
mass, even in a body supposed to be destitute of air and
water.

A genetic connection between the classes of volcanic
phenomena which have been mentioned is indicated by
many and various traces of simultaneity, and accompa-
nying transitions of the simpler and less powerful actions
into the more energetic and complicated ones. The
sequence in which I have arranged my materials, in the
mode of presentation which I have selected, is justified
by considerations of this nature. The magnetic activity
of our planet (of which the seat is not, indeed, to be
looked for in the molten interior, although, according to
Lenz and Bless, iron in a molten state is capable
of acting as the conductor of an electric or galvanic
current,) produces, when enhanced, a development of
light at the magnetic poles of the earth, or at least in
their vicinity. The first division of the present or tel-
luric volume of my work closed with this terrestrial lu-
minosity. I place at the opening of the second division,
or next in succession to this phenomena of a light-ex-
citing undulation of the ether caused by magnetic forces,
and first among volcanic phenomenon, that class of
volcanic activity which in its essential manifestations
acts only, as does the magnetic activity, dynamically :
exciting motions or undulations in the solid crust of the
earth, not producing or altering substances. Other not
essential, but in this point of view secondary, phenomena
(flames rising during earthquakes, eruptions of water,
and evolution of gas (228) following after the earthquake)
recall to the mind the phenomena of thermal waters

166 REACTION OF THE INTERIOR OF THE EARTH

and salses. Salsee sometimes show outbursts of flames,
visible to distances of very many miles, and rocks torn
from internal depths hurled into the air and scattered
around.(229) These, as it were, prepare the way for the
grand phenomena of volcanoes proper, which in turn,
in the intervals between distant epochs of eruption, ex-
hale, like salses, .only aqueous vapours and gases from
fissures. Thus striking and instructive are the analo-
gies presented in different stages by the gradations of
volcanism.

a. Earthquakes.

(Extension of the Representation of Nature, in Kosmos,
Bd. I. S. 210—225 ; Engl. Vol. I. p. 191—205.)

Since the publication of the first volume of the present
work (in 1845), and of the general description there
given of earthquake phenomena, the obscurity prevail-
ing respecting their seat and their causes has been but
little diminished ; but by the labours of (23°) Mallet
(1846) and Hopkins (1847) some valuable light has
been shed on the nature of the movement, the connection
of operations apparently diverse, and the separation to
be made between physical and chemical processes, asso-
ciated or occurring simultaneously. Here, as elsewhere,
advantage cannot fail to be derived from the application
of mathematical reasoning, as in the precedent set by
Poisson. In theoretical considerations on the dynamics
of earthquakes, the analogies between the undulations
of solid bodies, and the atmospheric waves of sound, to
which Thomas Young had called attention, (231) are
particularly adapted to lead to more simple and more
satisfactory views.

ON ITS EXTERIOR. EARTHQUAKES. 167

Change of place, agitation, uplifting, and the produc-
tion of fissures, mark the essential character of the
phenomenon. We must distinguish between the acting
force which gives the first impulse to the vibration, and
the constitution, propagation, and enhanced or diminished
intensity of the agitation-wave. In the " Representa-
tion of Nature," I described those effects which present
themselves immediately to the senses, and which I had
myself the opportunity of observing for so many years
on the sea, on sea-like plains (the llanos), on elevations
of from eight to sixteen thousand feet, on the margin of
the craters of active volcanoes, and in granitic and mica-
slate districts distant about twelve hundred geographical
miles from any fiery eruption, — in districts where at
particular periods the inhabitants no more count the
number of earthquake shocks than we in Europe count
the number of showers of rain, — where Bonpland and
myself, in journeying through the forest, had to dismount
from our mules on account of their becoming unruly
from uneasiness at the earth trembling uninterruptedly
beneath their feet for fifteen or eighteen minutes. By
being so long habituated to these phenomena, (an
advantage shared subsequently in a still higher degree
by Boussingault,) one becomes better disposed for calm
and accurate observation, for collecting and collating
with critical care, at the time and on the spot, various
and perhaps discordant accounts, and better prepared
for learning what have been the circumstances attendant
on great changes of the surface, of which fresh traces
are seen. Although, when we were on the spot, five
years had elapsed since the dreadful earthquake of
Riobamba, by which, on the 4th of February 1797,

168 REACTION OF THE INTERIOR OF THE EARTH

more than 30,000 human beings lost their lives in
the course of a few minutes, (232) we still saw the once-
moving cones of Moya (233) which had been then pro-
truded from the ground, and the combustible substance
of which they consisted employed in the huts of
the Indians for cooking. I could describe changes
of surface resulting from that catastrophe, quite analo-
gous, on a larger scale, to those presented by the cele-
brated Calabrian earthquake (February 1783), and the
accounts of which were long declared to be incorrect and
exaggerated, because not explicable by too hastily formed
theories.

Since, as I have already said, the considerations re-
specting the cause of the impulse to motion ought to be
carefully separated from those which relate to the nature
and the propagation of the waves of motion, the subject
is thereby divided into two classes of problems of very
unequal accessibility. In the first, as in so many cases
where we desire to ascend to the higher links in the chain
of causation, we cannot, in the present state of our
knowledge, look for any generally satisfactory results.
Nevertheless, while our efforts are directed to the dis-
covery of laws in phenomena which we are unable to
make the subject of actual observation, great cosmical
interest attaches to our at the same time keeping con-
stantly in view the different modes of genetic explanation
which have been put forward as probable. The greater
part of these (as in regard to all volcanic action or
volcanicity) relate, under various modifications, to the
high temperature and to the chemical constitution of
the molten interior of the earth ; one other most recently
proposed mode of explanation of earthquakes in trachytic

ON ITS EXTERIOR. EARTHQUAKES. 169

regions is furnished by geological conjectures respecting
the non-connection of volcanically upheaved masses of
rock. I have brought together the following brief no-
tices of the different views which have been taken of
the possible nature of the first impulse in earthquake
movements : — -

In one of these views the interior of the earth is
regarded as being in a state of igneous fluidity; — the
result of the process of formation of planets from a
gaseous state, whereby, in the transition from fluid to
solid, great disengagement of heat took place ; the
external strata, by radiation of their heat into space,
having been the first to cool down and solidify. The
unequal ascent of elastic vapours (formed at the limits
between the fluid and solid, either from the molten
mass of the earth only, or by the penetration of sea-
water), the sudden opening of fissures thereby occa-
sioned, and the sudden approach of these vapours,
formed at great depths, and having therefore great
heat and tension, to the higher rocky strata nearer
to the earth's surface, — are supposed to occasion agi-
tation, and to give the first impulse to the earthquake-
wave. An auxiliary influence, due to a non-telluric
cause, is supposed to exist in the attraction of the
moon and sun (234) acting on the fluid molten surface
of the earth's nucleus, whereby there must arise in-
creased pressure, — either immediately, against a solid
superincumbent rocky vault, — or mediately, where in
subterranean basins the solid is separated from the
liquid molten mass by elastic vapours.

In another view the nucleus of our planet is regarded

170 REACTION OF THE INTERIOR OF THE EARTH

as consisting of unoxidised masses of the metallic bases
of earths and alkalies ; and it is supposed that water
and air gain access to these and thus that volcanic
activity is excited. Volcanoes do, indeed, send forth
great quantities of aqueous vapour into the atmo-
sphere ; but the hypothesis of a penetration of water
to the volcanic hearth would have to encounter great
difficulties in respect to the mutual pressure (235)
of the external column of water and the internal lava ;
and the absence during eruptions, or at least the
great rarity of ignited hydrogen, which is not ade-
quately replaced by the formations of hydrochloric
acid, ammonia., and sulphuretted hydrogen, (236) led
the illustrious originator of this hypothesis sponta-
neously to relinquish it.(237)

According to the third view, which is that of the ac-
complished South-American traveller, Boussingault,
a want of continuity in the masses of trachyte and
dolerite, of which the upheaved volcanoes of the chain
of the Andes are composed, is regarded as^ a principal
cause of many and very widely acting earthquakes in
those regions. The colossal cones and dome-shaped
summits of the Cordilleras are in this view supposed
to have been by no means upheaved in a state of soft-
ness or semi-fluidity, but are regarded as consisting of
enormous angular fragments, which have been pushed
up and heaped one upon another. In the course of
these operations it would necessarily have followed
that, in the vast piles so formed, great intervals and
cavities would have occurred, and that by subsidence,
and by the occasional fall of imperfectly supported
masses, commotions would ensue. (238)

ON ITS EXTERIOR. EARTHQUAKES. 171

With greater clearness of view than is attainable in
considerations respecting the nature of the first impulse
(which, indeed, may be supposed to be of different kinds),
we are able to reduce the action originated by it, i.e.
the undulations of the earthquake-waves, to simple me-
chanical theories. As I have already remarked, this
part of our knowledge of nature has latterly been very
materially augmented. The undulations have been de-
scribed in their progress and in their propagation through
rocks of different degrees of density and elasticity ; (239)
and the causes of the diminution in their velocity of
propagation by the occurrence of interruptions, reflexions,
and interferences of waves (24°) have been mathematically
investigated. The apparently rotatory (or " vorticose ")
shocks, of which the obelisks in front of the convent of
San Bruno, in the little town of Stephano del Bosco m
Calabria (1783) present an example which has been
much discussed, have been attempted to be reduced to
rectilinear shocks. (241) Waves or undulations of air, of
water, or of earth, do, indeed, all follow as respects
space the same known dynamical laws ; but the earth-
waves are accompanied in their devastating action by
phenomena which by their nature remain enveloped in
greater obscurity, and belong to the class of physical
processes. Such are exhalations or outpourings of
vapours in a state of tension, gases, or (as in the small
moving Moya-cones of Pelileo,) gritty mixtures of py-
roxene crystals, carbonaceous matter, and silicious-
shelled infusoria. These moving cones have overturned
many Indian huts.(242)

In the general Description of Nature in the first
volume, in connection with the great catastrophe of

172 REACTION OF THE INTERIOR OF THE EARTH

Riobamba, Feb. 4, 1794, many facts were collected by
me on the spot from the lips of survivors, and with
the most solicitous desire to elicit historic truth. Some
of these, as has been already stated, were analogous
to circumstances which took place in the great Cala-
brian earthquake of 1783 ; others, which were new,
were especially characterised by an explosive action from
below upwards. The earthquake itself was neither an-
nounced nor accompanied by any subterranean noise :
but a prodigious noise, still designated simply as " el gran
ruido," was first heard eighteen or twenty minutes later,
and only under the two towns of Quito and Ibarra, at a
distance from Tacunga, Hambato, and the chief theatre
of devastation. In the history of catastrophes suffered
by man, there is no other instance in which in the
course of a few minutes, and in a thinly peopled moun-
tainous country, so many thousand lives were lost by
the production and passage of a few earth-waves, accom-
panied by the opening of fissures. In reference to this
earthquake, of which the first accounts were given by
the celebrated Valencia botanist, Don Jose Cavanillas,
particular attention is further due to the following
phenomena : — Fissures, which alternately opened and
closed, so that persons partially engulfed were saved by
extending their arms that they might not be swallowed
up ; portions of long trains of muleteers and laden mules
(recuas), disappearing in suddenly opening cross fissures,
whilst other portions, by a hasty retreat, escaped the dan-
ger ; vertical oscillations, by the non-simultaneous rising
and sinking of adjoining portions of ground, so that
persons standing in the choir of a church, sixteen feet
above the pavement of the street, found themselves

ON ITS EXTERIOR. EARTHQUAKES. 173

lowered to the level of the pavement without being
thrown down; the sinking down of massive houses, (243)
with such an absence of disruption or dislocation, that
the inhabitants could open the doors in the interior, pass
uninjured from room to room, light candles, and debate
with each other their chances of escape, during two days
which elapsed before they were dug out ; lastly the en-
tire disappearance of great masses of stones and building
materials. The old town had possessed churches, con-
vents, and houses of several stories, but in the places
where they had stood, we found, on tracing out among
the ruins the former plan of the city, only stone heaps,
of from eight to twelve feet high. In the south-western
part of old Eiobamba (in the former Barrio de Sigchu-
guaica), we could distinctly recognise the signs of an
explosion from beneath, the action of a force from
below upwards. On the cone called Cerro de la Culca,
which rises some hundred feet and appears above and
on the north side of the Cerro de Cumbicarca, rubbish
from the stone buildings lay mingled with the ske-
letons of men. Movements of translation in a hori-
zontal direction, whereby avenues of trees were re-
moved without being uprooted, or fields, bearing
different kinds of . cultivation, became intermixed,
showed themselves repeatedly, both in Quito and in
Calabria. A still more striking and complicated phse-
nomenon was the finding articles belonging to one house
among the ruins of others at a considerable distance —
a discovery which gave rise to some lawsuits. Is it, as
the inhabitants of the country believe, that the earth
throws out again at one spot that which it has swallowed
up at another ? or, is it (notwithstanding the distance)

174 REACTION OF THE INTERIOR OF THE EARTH

a simple transport over the earth's surface ? As in
nature effects are reproduced on the return of similar
conditions, we ought unhesitatingly to mention even
incomplete observations in order to direct the attention
of future observers to special phenomena.

It is not to be forgotten that in most cases of the
production of fissures, the earth-wave or undulation, by
which the solid parts are agitated, is associated, accord-
ing to my experience, with a concurrent action of very
different forces, manifested in emanations of gases and
vapours. If, in the undulatory movement, the extreme
limit of the elasticity of the material (of the different
kinds of rock or loose earthy strata) is overpassed, and
rupture or separation ensues, elastic fluids in a state of
tension may escape through the openings thus formed,
bringing from the interior to the surface different sub-
stances, whose outburst may again become in turn a
cause of fresh translatory movement. To the class of
these merely accompanying phenomena of the primary
agitation or earthquake, belongs the elevation of uncon-
testedly moving Moya cones; and, probably, also the
transport of objects on the surface of the earth. (244)
When, in the formation of great fissures, they close again
in their upper portions only, the subterranean cavities
thus produced may not only become the cause of fresh
earthquakes, — inasmuch as, according to Boussingault's
conjecture, ill-supported masses become in time detached,
and, in sinking, cause commotion, — but we may also
conceive it possible that in the fissures opened by earlier
earthquakes, elastic fluids may become active in occa-
sioning fresh earthquakes in places to which they for-
merly had no access. It would thus be the accompanying

ON ITS EXTERIOE. EARTHQUAKES. 175

or secondary phenomenon, and not the intensity of the
undulation which had traversed a portion of the
earth's crust, which would be the true occasion of the
gradual, very important, and too little regarded exten-
sion of circles of commotion, or earthquake regions. (245)
Manifestations of volcanic activity, to the lower gra-
dations of which earthquakes belong, almost always
comprise at once phenomena of movement, and physical
production of substances. I have already noticed in
the Description of Nature in the first volume, how
there rise from fissures at a distance from any volcano : —
water and hot steam ; carbonic acid gas and other
suffocating exhalations ; black smoke, (as, for many
days, at the rock of Alvidras in the Lisbon earthquake
of Nov. 1, 1755); flame, sand, mud, and Moya with
intermixed carbonaceous matter. The great geologist
Abich has pointed out the connection between the ther-
mal springs of Sarcin (about 5382 feet high), in Grhilan
in Persia, on the road from Ardebil to Tabreez, and the
earthquakes by which those highlands are visited at
intervals of two years. In October 1848, an undulatory
movement of the ground, which lasted an entire hour,
obliged the inhabitants of Ardebil to leave the town,
and at the same time the temperature of the springs,
which is between 44° and 46° Cent. (111-2° and 114-8°
Fahr.), rose for a whole month to scalding heat.(24G)
Perhaps there is no place on the earth where, as Abich
has told us, " the intimate connection of earthquakes
producing fissures, with the phenomena of mud vol-
canoes, salses, inflammable gases permeating through
loose soil, and petroleum springs, is more distinctly
indicated, and may be more clearly recognised, than at

176 REACTION OF THE INTERIOR OF THE EARTH

the south-eastern extremity of the Caucasus between
Shemacha, Baku, and Sallian. It is that portion of the
great Aralo-Caspian depression in which the ground is
most frequently agitated." (247) When in Northern Asia,
I was myself struck by the circumstance that the circle
of commotion, or earthquake region, of which the
district of Lake Baikal appears to be the centre, extends
to the westward only as far as the easternmost part
of the. Eussian Altai, as far as the silver mines of
Eiddersk, the trachytic rock of Kruglaia Sopka, and
the hot springs of Eachmanowka and Arachan, but not
to the chain of the Oural. Farther to the south,
beyond the parallel of 45° lat., there is found in the
chain of the Thian-schan, a zone running east and west,
characterised by every kind of manifestation of volcanic
activity. Not only does it extend from the " fire-dis-
trict " (Ho-tscheu) in Tourfan through the small Asferah
chain to Baku, and from thence past Mount Ararat to
Asia Minor, but it is even thought that it may be traced
between the latitudes of 38° and 40°, through the vol-
canic basin of the Mediterranean to Lisbon and the
Azores. I have treated this important subject of vol-
canic geography in detail elsewhere.(248) In Greece also,
which has suffered more than any other part of Europe
from earthquakes (Curtius, Peloponesos, Bd. i. S. 42 —
46), a countless number of warm springs, either still
flowing or which have disappeared, have broken forth
during earthquake shocks. A thermic connection of
this kind was already pointed out in the remarkable
book of John Lydus on earthquakes. (De Ostentis, cap.
liv. p. 189, Hase.) The great natural event of the de-
struction of Helice and Bura, in Achaia (373 B. c.),

ON ITS EXTERIOR. EARTHQUAKES. 1/7

gave especial occasion to the formation of hypotheses
respecting causal connection in volcanic activities. It
was then that Aristotle originated the singular theory of
the power of winds entangled in the bowels of the earth.
(Met. II. p. 368.) The frequency of earthquakes in
Greece and Lower Italy, by destroying the monuments
of the most nourishing period of Art, has exercised a
most unfavourable influence on the study of the Greek
and Roman cultivation at different periods. Egyptian
monuments also (for example a colossal statue of Mem-
non, 27 B. c.) have suffered from earthquake shocks,
which, as Letronne has shown, were not so rare in the
Valley of the Nile as has been supposed. (Les Statues
Vocales de Memnon, 1833, p. 23-27, and 255.)

The considerations on which we have been dwelling
will make it appear still more remarkable, that so many
warm sanatory springs should have retained their com-
position and their temperature unaltered for centuries ;
and must therefore flow from fissures which have appa-
rently undergone no changes either at their bottom or
sides. Channels of communication opened with higher
strata would have diminished, and with lower strata
would have increased, the temperature of the spring.

At the great outburst of the volcano of Conseguina in
the State of Nicaragua, on the 23rd of January 1835,
the subterranean noise (249) (los ruidos subterraneos)
was heard simultaneously in the island of Jamaica and
on the highlands of Bogota (nearly 9000 feet above the
sea) ; a distance greater than that between Algiers and
London. I have already remarked elsewhere that at
the eruption of the volcano in the Island of St. Vincent,
on the 30th of April 1812, at two in the morning, a
VOL. iv. N

178 REACTION OF THE INTERIOR OF THE EARTH

noise similar to a cannonade, unaccompanied by any sen-
sible earthquake movement, was heard over a space of
10,000 (German) geographical square miles, 15 to a
degree. (25°) It is well deserving of notice that when
earthquakes are associated with subterranean noise,
which is by no means always the case, the loudness of
the noise does not increase with the 'strength of the
earthquake. The most singular and wonderful pheno-
menon of the production of sound of which we have any
record is still the subterranean and roaring <s bramidos de
Gruanaxuato," from the 9th of January to the middle
of the following month in 1784 ; respecting which I
was able to obtain the first well assured intelligence
from the lips of living witnesses, and from original offi-
cial documents. (Vol. I. p. 196, and note 187.)

The velocity of earthquake propagation at the surface
must be supposed to undergo a considerable modifica-
tion from the very different densities of rocks, (granite
and gneiss, basalt and trachytic porphyry, Jurassic
limestone and gypsum,) and of alluvial soil or detritus,
passed through by the undulation. It would, however,
be exceedingly desirable to learn something as to the ex-
treme limits between which the rates of velocity vary.
It is probable that the most violent earthquakes are by
no means always the most rapidly propagated. The
measurements, moreover, do not always apply to the path
which the earthquake-waves may have followed. Ac-
curate mathematical determinations are greatly wanted.
Quite recently, a result has been obtained with great
exactness and judicious consideration of all circum-
stances, by Julius Schmidt, assistant at the Astrono-
mical Observatory of Bonn, on the Rhine earthquake

ON ITS EXTERIOR. EARTHQUAKES. 179

of the 29th of July 1846. The velocity of propagation
obtained for that earthquake was 1466 English feet in
a second. This is a rapidity exceeding that of the -wave
of sound in the air ; but if we consider that the velo-
city of the sound-wave in water is, according to Colladon
and Sturm, 5016 English feet, and in cast-iron tubes,
according to Biot, 11,393 feet in a second, the result
found for the earthquake-wave will appear to us very
small. For the Lisbon earthquake of Nov. 1, 1755,
from the coast of Portugal to that of Holstein, Schmidt
found (from less accurate data) a velocity more than
five times greater than in the case of the Ehine earth-
quake of the 29th of July 1846. Between Lisbon and
Grluckstadt (a distance of 1180 English geographical
miles), the rate derived by him is 7955 English feet in
a second ; which is still 3438 feet less than takes place
in cast iron. (251)

Earthquakes, and sudden fiery eruptions of volcanoes,
recurring after long intervals of repose, — whether they
send forth merely scoriae, or, like intermitting water-
springs, pour forth fluid molten earths in lava streams,
— have indeed the same causal connection, or common
origin, in the high temperature of the interior of our
planet ; but one of these phenomena shows itself in most
cases quite independently of the other. For example,
violent earthquakes, in their linear extension along the
chain of the Andes, agitate regions in which there are
volcanoes which are not extinct, and are indeed often in
a state of activity, yet without causing in them any per-
ceptible excitement. In the great catastrophe of Eio-
bamba, the neighbouring volcano of Tunguragua and the
rather more distant one of Cotopaxi remained quite tran-

180 REACTION OF THE INTERIOR OF THE EARTH

quil ; and inversely, volcanoes have had great and long-
continued eruptions without earthquakes having been felt
in the country around, either previously or at the same
time. The most devastating earthquakes recorded in
history, and which have passed through very many
thousands of square miles, are precisely those which, so
far as could be perceived at the surface of the earth, had
no connection with the activity of volcanoes. This class
has latterly been termed plutonic, — in opposition to
properly volcanic earthquakes which are mostly limited
to lesser areas. This nomenclature, which would require
far the greater number of earthquakes to be called
plutonic, is not to be approved as respects more ge-
neral views.

We have everywhere beneath our feet that which may
produce earthquake shocks; and the consideration that
almost three fourths of the earth's surface are covered by
sea, and (omitting a few sporadically scattered islands)
without any permanent communication between the in-
terior and the atmosphere (i.e. without active volcanoes),
sufficiently refutes the erroneous but widely-spread
belief, that all earthquakes are to be attributed to
the eruption of a distant volcano. Continental earth-
quakes are indeed often propagated over the bottom of
the sea, and the agitation communicated from the
coast, causes those formidable sea-waves of which the
earthquakes of Lisbon, Callao, and Chili have given
such memorable examples. When, on the other
hand, the earthquake proceeds from the bottom of the
sea, from the dominions of the earth-shaking Poseidon
(cn-iaixdaiv, Kiin}(jiyQ<£>v), and is not accompanied by an
upheaval producing an island (as in the ephemeral

ON ITS EXTERIOR. EARTHQUAKES. 181

existence of the Island Sabrina or Julia), there may be
remarked at points where yet the mariner would not
feel any shock, an unusual heaving and rolling of the
waves. The inhabitants of the Peruvian coast often
directed my attention to this phenomenon. In the
harbour of Callao, and at the Island of San Lorenzo
which is opposite to it, in perfectly calm, windless
nights, in this exceedingly tranquil part of the Pacific,
I have myself seen suddenly wave rise upon wave to a
height of eleven to fourteen feet, continuing for a few
hours. Nor can we, in these latitudes, as we might do
elsewhere, explain the phenomenon by the supposition
of a violent storm having occurred far out at sea.

To begin with agitations of the earth limited to the
smallest space, and obviously owing their origin to the
activity of a volcano, I will mention first what I ob-
served when seated with my chronometer in my hand
at night in the crater of Vesuvius, at the foot of a
small cone of eruption; it was after the great earth-
quake at Naples, on the 26th of July 1805, and after
the eruption of lava which followed seventeen days
later. I regularly felt the ground of the crater shake
every twenty or twenty-five seconds, immediately before
each eruption of glowing scoriae or cinders. Of these,
which were thrown to a height of fifty or sixty feet, part
fell back into the opening from which they had issued,
and part covered the sides of the cone. The regularity
of the phaenomenon rendered the observation free from
danger. The repeated small shocks which I felt were
not sensible beyond the crater, i. e. at the Atrio del
Cavallo, or at the hermitage del Salvatore. The pe-
riodicity of the shaking showed that it' depended on

182 REACTION OF THE INTERIOR OF THE EARTH

a definite degree of tension, which it was necessary that
the vapours should reach before they could burst through
the molten mass in the interior of 'the cone of cinders.
Just as in the case described, in which no shocks were
felt on the sides of the cinder-cone of Vesuvius, so in
a quite analogous phenomenon, though on a far grander
scale, on the cone of ashes of the volcano of Sangai, which
rises south east of the city of Quito to a height of
17,100 English feet, a very distinguished observer, Herr
Wisse, having (in December 1849) approached within a
thousand feet of the summit and crater, perceived no
quaking of the ground, (252) although 267 explosions
(eruptions of scoriae) were counted in the course of a
single hour.

A second, immensely more important, class is the very
numerous one which consists of earthquakes usually
accompanying or preceding great eruptions, whether of
volcanoes which, like our European ones, pour forth
streams of lava ; or of volcanoes which, like Cotopaxi,
Pichincha, and Tunguragua in the Andes, send forth
only scoriae, ashes, and vapours. It is especially in
regard to this class that volcanoes may be viewed as
safety-valves, or vents, according to the early remark
of Strabo on the fissure from which lava flowed, near
Lelante, in the island of Eubcea. The earthquakes
cease when a considerable eruption has taken place.

The most wide-spread devastations (253) are those occa-
sioned by earthquake-waves which traverse partly non-
trachytic and non-volcanic countries, and partly trachytic
and volcanic ones, as the Cordilleras of South America
and Mexico, without exercising any influence on the
neighbouring volcanoes. These form the third class or

ON ITS EXTERIOR. EARTHQUAKES. 183

group of phenomena, and it is that which points most
strongly to the existence of a general cause in the
thermic constitution of the interior of our planet. To
this third group belongs also a case of rare occurrence
in which, in countries non- volcanic and rarely visited by
earthquakes, the ground trembles uninterruptedly for
several months, on a very restricted space, seeming to
presage an upheaval, and the formation of an active vol-
cano. This took place in the beginning of the present
century in the Piedmontese valleys of Pelis and Clusson,
as well as at Pignerol in April and May 1808, and also
in the spring of 1829, in Murcia, between Orihuela and
the sea-coast, on a space rather less than a (German)
square mile. When in the interior of Mexico, on the
western slope of the high land of Mechoacan, the cul-
tivated flat of Jorullo was incessantly shaken for ninety
days, the volcano rose surrounded by many thousand
small cones, about five or seven feet high (los hornitos),
and poured forth a brief but powerful stream of lava.
On the other hand, in Piedmont and in Spain, the
shaking of the earth gradually ceased without any great
natural event ensuing.

I have thought it useful to enumerate the wholly
different kinds of manifestation of the same volcanic
activity (or reaction of the interior of the earth on
its exterior), in order to guide the observer, and*aid in
providing materials which may lead to fruitful results
respecting the causal connection of the phenomena.
Sometimes the volcanic activity embraces at one time,
or at times little distant from each other, so large a
portion of the earth, that the shocks which are felt may
be attributed to several kindred causes at once. The

184 REACTION OF THE INTERIOR OF TEE EARTH

years 1796 and 1811 especially offer remarkable ex-
amples (254) of such a grouping of phenomena.

b. Thermal Springs.

(Enlargement of the Picture of Nature in Cosmos, Vol. I.
p. 207—211.)

We have described Earthquakes as a result of the
vital activity of the interior of our planet, manifesting
itself in irregularly repeated and often fearfully de-
structive phenomena. In earthquakes, regarded accord-
ing to their essentially inherent character, the volcanic
power displayed acts dynamically only, or as an energy
giving rise to motion ; although in some localities, under
particular collateral conditions, it may be the means, not
indeed of producing particular kinds of substances (as
do volcanoes proper), but of bringing them to the
surface of the earth. In like manner as in earthquakes,
water, vapours, petroleum, mixtures of gases, or pasty
masses (mud, and "moya") are occasionally, and
during periods of brief duration, brought to the surface
and emitted through suddenly opened fissures; — so
fluids, both liquid and gaseous, flow permanently from
the bosom of the earth through the existing generally
distributed network of intercommunicating fissures.
We tlfcus place by the side of the transitory and tumul-
tuous phenomena of emission, that vast and tranquil
system of springs, wells, and fountains, by which organic
life is beneficently nourished and refreshed, and by
means of which, for thousands of years, the moisture
withdrawn from the atmosphere by the fall of rain, is
restored to the service of the organic creation. Analo-
gous phenomena in the great economy of nature are

ON ITS EXTERIOR. THERMAL SPRINGS. 185

mutually illustrative of each other, and some allusion
to such connecting links as can be recognised between
them should not be omitted where generalisation of
view is desired.

The division of springs into warm and cold, which is
in such general use, and in ordinary parlance appears so
natural, will be seen to rest only on very uncertain
foundations, if it is wished to reduce it to definite nu-
merical expression. If the temperature of springs is
to be compared with that of the human body (the in-
ternal heat of which Brechet and Becquerel found with
thermo-electric apparatus to be from 98-1° to 98'6°Fahr.)
the degree of the thermometer scale at which a liquid
when placed in contact with parts of the body would be
called cold, warm, or hot, will differ considerably ac-
cording to individual feelings. No absolute degree of
temperature can be determined on, as that above or
below which a spring shall always be termed warm or
cold. The proposal to call, in each climatic zone, a
spring, whose mean annual temperature does not ex-
ceed the mean annual temperature of the air in the
same zone, a cold spring, offers, at least, scientific
accuracy in the comparison of definite numbers. It has
also the advantage of leading to considerations respecting
the different origin of springs; since the agreement
with the mean annual atmospheric temperature is to be
looked for in "invariable" springs directly, and in
" variable " ones (as Wahlenberg and the elder Erman
have shown) in the mean of their summer and winter
temperatures. But according to this criterion, in one
zone a spring may be termed " warm," whose tempera-
ture is scarcely a seventh or an eighth part of that of
another which being situated in the equatorial regions

186 REACTION OF THE INTERIOR OF THE EARTH

is termed " cold." I take the difference between the
mean temperatures of St. Petersburg (38'1° Fahr.),
and the banks of the Orinoco. The purest spring
waters that I ever drank while I remained in the
district of the Cataracts of Atures and Maypures, (255)
or in the forests of Atabapo, had a temperature of up-
wards of 26° Cent, or 78-8° Fahr. And the tempera-
ture of the great rivers of tropical South America
corresponds to that of these which would in such case
be technically called " cold " springs 1 (256)

The flowing forth of springs, occasioned by various
causes of pressure, and by the interconnection of fissures
containing water, is a phenomenon so general over the
earth's surface, that at some points it takes place in the
highest uplifted mountain strata, and at others at the
bottom of the sea. In the first quarter of the present
century, Leopold von Buch, Wahlenberg, and I collected
numerous results respecting the temperature of springs
and the distribution of heat in the interior of the earth
in both hemispheres, from the 12th degree of south to the
71st degree of north latitude. (257) The springs which
have an " invariable " temperature were carefully dis-
tinguished from those which vary with the seasons ; and
Von Buch remarked the powerful influence of the dis-
tribution of the fall of rain in the course of the year,
or the relative frequency and abundance of winter-
and summer-rains, on the temperature of the "vari-
able springs," which are, generally speaking, the most
numerous and widely diffused class of springs. This
influence has recently been placed in a clearer light,
both in a geographical and hypsometrical view (or
according to latitude and to elevation), by de Gasparin,

ON ITS EXTERIOK. THERMAL SPRINGS. 187

Schouw, and Thurmann. (258) Wahlenberg maintained
that in very high latitudes the mean temperature of
variable springs is somewhat higher than the mean
temperature of the air ; he looked for the cause of this,
not in the dryness of very cold air and the consequent
less abundance of winter rain, but in the protecting
covering of snow diminishing the radiation of heat
from the ground. In those parts of the great plains of
Northern Asia in which a perpetual " stratum of ice,"
or at least frozen soil with interspersed pieces of ice, is
found at a depth of a few feet, (259) the temperature of
springs cannot, without great caution, be used in making
out Kupffer's important theory of isogeothermal lines.
In such situations there takes place a twofold radiation
of heat from the upper stratum of the earth ; upwards
towards the atmosphere, and downwards towards the ice
stratum. An extensive series of valuable observations
collected by my friend and companion Grustav Eose, in
the Siberian Expedition, in the heat of summer (often
in wells still surrounded with ice), between the Irtysch,
the Obi, and the Caspian Sea, showed a great complica-
tion of local disturbances. Similarly perturbing in-
fluences, arising from entirely different causes, show
themselves in results obtained within the tropics, from
mountain springs issuing forth on vast table-lands eight
or ten thousand feet above the sea (Micuipampa, Quito,
Bogota), or on slender isolated summits many thousand
feet still higher; these comprehend a more extensive
part of the earth's surface, and are useful in leading
to the consideration of analogous thermic relations in
mountainous countries in the temperate zone.

In this important subject it is first of all necessary to

188 REACTION OF THE INTERIOR OF THE EARTH

separate actual observations from the theoretical con-
clusions which have been based upon them. The ob-
ject we aim at, enounced in its greatest generality, is
a threefold one: — the distribution of heat, in the ac-
cessible part of the earth's crust, in the aqueous cover-
ing (the ocean), and in the atmosphere. In the liquid
and gaseous coverings of the globe, an opposite variation
of temperature takes place in the vertical direction.
In the solid parts of the globe, the temperature in-
creases with increasing depth ; the variation being here
the same in direction (though the ratio is very different)
as in the atmospheric ocean, in which elevated plateaus
and variously shaped mountain summits represent shoals
and submarine rocks. We are acquainted by direct
experiment with the distribution of heat in the at-
mosphere, geographically or in different latitudes and
longitudes, and hypsometrically or at different amounts
of vertical elevation above the level of the sea. This
knowledge is, however, almost wholly confined to the
near vicinity of the solid or liquid surface. Scientific
and systematically arranged investigations, by means of
aerostatic explorations of the aerial ocean beyond the
immediate proximity of the disturbing influences of the
earth, are still so much too few in number, that they
have for that reason been little capable of affording the
requisite numerical data for mean conditions. In re-
gard to the decrease of temperature in the depths of
the ocean, observations are not wanting; but currents
bringing water from different latitudes and depths, and
of different densities, interpose almost greater obstacles
to the attainment of general results than do the currents
of the atmosphere. The thermic conditions of the two

ON ITS EXTERIOR. THERMAL SPRINGS. 189

coverings of our planet, which will he treated of more
particularly in the sequel, are here referred to in pass-
ing, in order that the influence of the vertical distri-
bution of temperature in the solid crust of the earth,
(the system of " geoisothermals,") may not be looked
at in too isolated a manner, but that it may be regarded
rather as a portion of the all-pervading movement of
heat, — a true " cosmical activity."

Instructive as are in many respects observations on
the unequal decrease of temperature with increasing
elevation of their point of issue of such springs as do
not vary with the seasons, yet the local law of such de-
crease ought not to be regarded, as it too often is, as a
general "geothermic" law. If we were assured that
the springs came from unmixed waters, resting on very
extensive horizontal strata, we might indeed believe
them to have gradually acquired the temperature of the
solid materials in the vicinity ; but the great network
of fissures by which the upheaved masses are inter-
sected must render this only a rare case. Colder waters
from higher levels mix with the lower ones. Our
mining operations, small as is the depth to which they
penetrate, are very instructive in this respect ; but we
shall not arrive directly at the knowledge of the geo-
thermal lines until observations are obtained at very
various elevations above the level of the sea by Bous-
singault's method, i.e. by sinking the thermometers
into the earth below the depth to which the influence
of the variations of the adjacent atmosphere extends.(260)
From the 45th parallel of latitude to the parts of
the tropics nearest to the equator, the depth at which
the " invariable " stratum begins decreases from 64 feet

190 KEACTION OF THE INTERIOR OF THE EARTH

to about 2 feet. It is therefore only within the tropics,
or at no great distance beyond them, that this method
of arriving at the knowledge of the mean temperature
of the ground is easily practicable. The valuable re-
source offered by Artesian wells (which, with absolute
depths of from 750 to 2340 feet in round numbers,
have shown an increase of temperature of 1° Fahr. for
a mean descent of 56 feet,) has as yet only been
available to physical inquirers in districts not more
than 1600 feet above the level of the sea. (261) I have
visited excavations made in search of silver ores in the
Andes 6° 45' south of the equator, at an elevation of
13,200 feet, and I there found the temperature of the
waters oozing out of the fissures in the limestone rock
52-3° Fahr.(262) The waters which on the ridge of the
Andes, at the Paso del Assuay, were warmed for the
baths of the Inca Tupac Yupanqui, came probably from
the springs of the Ladera de Cadlud ; where, by baro-
metric measurement, I found the elevation of the path
(near which ran also the ancient skilfully constructed
Peruvian road,) 15,525 feet above the sea, or nearly
equal to the height of Mont Blanc.(263) These are the
highest points at which I was ever able to observe
spring water in South America. In Europe, in the
Eastern Alps, the brothers Schlagintweit, at the height
of 9440 feet, found the temperature of the water in the
gallery of a gold mine (the " Groldzeche "), and of small
springs near the mouth of the gallery, only 33*4° Fahr.,(264)
at a distance from either snow or glacier ice. The
limits of elevation at which springs are met with differ
with differences of latitude and of the height of the
line of perpetual snow ; and also with the different pro-

ON ITS EXTERIOR. THERMAL SPRINGS. 191

portions which the elevations of the loftiest peaks bear
to those of the general crests of the mountains and of
the table-lands.

If the semi-diameter of our planet were increased by
the height of the Himalayan summit of Kinchinjunga,
28-175 feet, this addition of only about -gj-g- of the earth's
radius would scarcely make any difference (according
to Fourier's analytic theory) in the temperature of the
surface as cooled by radiation; the thermal condition of
the upper portion of the crust of the earth would be
almost exactly the same as it now is. But where par-
ticular parts of the surface rise into the atmosphere,
whether in mountain chains, or in isolated summits,
there takes place in the interior of the upheaved strata a
decrease of warmth from below upwards, which decrease
is modified by contact with strata of air of unequal
temperatures; by the different capacity for heat and the
different heat-conducting power of different kinds of
rock ; by the exposure to the sun's rays of the bare or
forest-covered summits and declivities ; and by the
greater or less radiation from the mountain, according
to the particular form of its relief, w'hether forming a
wide mass, or one or more narrow cones or pyramids.
The particular height of the region of clouds; the
extent of icy or snowy covering (varying according to
the different heights of the snow limit); and the
frequency of the currents of colder air, which at par-
ticular parts of the twenty-four hours roll down the
steep declivities; all modify the effects of terrestrial
radiation. As the soaring jagged peaks become colder,
there arises in the interior a faint current of heat from
below upwards, tending towards, but never attaining,

192 REACTION OF THE INTERIOR OF THE EARTH

the production of a state of equilibrium. The recog-
nition of so many agencies affecting the -vertical
distribution of heat, leads to well-founded conjectures
respecting the connection of complicated local pheno-
mena, but not to the derivation of any direct numerical
determinations. In mountain springs, (and the higher
ones, being important to the Chamois hunters, are
carefully sought out,) a doubt often subsists as to their
being mixed, either with waters which sinking down
from above bring a colder temperature, or with waters
rising from below and bringing with them the higher
temperature of more deeply seated strata. From
nineteen springs observed by Wahlenberg, Kamtz
draws the conclusion that in the Alps we require to
ascend between 960 and 1020 feet, to see the tempe-
rature of springs diminish 1*8° Fahr. (average, 550 feet to
1° Fahr.) More numerous and more carefully selected
observations by Hermann and Adolph Schlagintweit, in
the eastern Carinthian, and the western Swiss Alps round
Monte Rosa, give a more rapid decrease of temperature,
viz. 1-8° Fahr. to only 767 feet (426 feet to 1° Fahr.).
According to the great work of these excellent ob-
servers, (265) "the rate of decrease of the temperature
of the springs with increasing elevation is in every case
somewhat slower than that of the mean annual tem-
perature of the atmosphere, which in the Alps gives
319 English feet for l°Fahr. The springs there are,
generally speaking, warmer than the mean tempera-
ture of the air at the same level; and this difference
between the temperature of the air and the springs
increases with the elevation. The temperature of the
ground at equal heights is not the same throughout

ON ITS EXTERIOR. THERMAL SPRINGS. 193

the Alps ; for, apart from the influence of geographical
latitude, the elevation of the isothermal planes, con-
necting the points at which the springs have the same
mean temperature, increases with the increasing mean
elevation of the surrounding country: all in accordance
with the laws of the distribution of heat in a solid body
of varying thickness, to which the relief (or mass ele-
vation) of the Alps may be compared."

In the Andes, and precisely in that volcanic portion
of the chain which presents the greatest elevations, the
sinking of thermometers in the ground may in some
cases, under the influence of particular circumstances,
give rise to delusive results. In pursuance of a pre-
viously formed opinion, that the entire absence of snow
on the dark ridges of black rock, seen from afar inter-
secting the snowy region, may not always be exclusively
due to the configuration and steepness of their sides,
but may also be in part attributable to other causes, —
I plunged the bulb of a thermometer only three or four
inches into the sand, which filled the cleft in such a
ridge on the Chimborazo, at an elevation of 18,288
feet, or 2550 feet higher than the summit of Mont
Blanc. The thermometer showed persistently 42° -4
Fahr., while the external air was only 36°*9. The result
of this observation is of some importance, because 2560
feet lower down, at the lower limit of perpetual snow
on the volcanoes of Quito, the mean temperature of the
air, inferred from a large number of observations col-
lected by Boussingault and myself, is under 35°. The
temperature of 42°-4 must, therefore, have been due to
the subterranean warmth of the dolerite mountain :
I do not mean to say to the temperature of the entire

YOL. iv. O

194 REACTION OF THE INTERIOR OF THE EARTH

mass, but to that of currents of air rising from lower
depths. There is, moreover, at the foot of the Chimbo-
razo, at an elevation of 9500 feet, not far from the
village of Calpi, a small crater of eruption called
Yana-Urcu, which, in the middle of the 15th century,
appears to have been in a state of activity, as is also
indicated by its black scoriaceous rock (augitic por-
phyry.) (»•)

The aridity of the plain out of which the Chimborazo
rises, and the existence of the subterranean brook
of which the rushing sound is heard under the above-
named volcanic hill Yana-Urcu, led both Boussingault
and myself, (267) at very different times, to consider that
the waters proceeding from the daily melting of the
enormous masses of snow at their lower limit, sink down
into deep clefts and hollows below the upheaved volca-
noes. These waters produce continual refrigeration
in the strata through which they sink. But for this
agency the entire -dolerite and trachyte mountains, even
at times when there is no indication of any proximate
eruption, woidd have a still higher internal temperature
imparted to them by the ever-acting primeval volcanic
source of heat, which, perhaps, is not at an equal depth
in all zones of the earth. Thus, in the conflict between
heating and cooling causes, the existence of a perpetual
flow of heat both upwards and downwards may be most
particularly assumed where the solid parts of the earth's
surface rise high into the atmosphere.

Mountains and high summits, however, regarded
relatively to the area which they occupy, are but a very
small phenomenon in the relief-form (or vertical confi-
guration of the dry land) ; and, moreover, almost two

Oft ITS EXTERIOR. THERMAL SPRINGS. 195

thirds of the whole surface of the earth (according
to the present state of geographical discovery in the
polar regions of either hemisphere, the ratio of sea and
land may be taken as 8:3) are sea-bottom. This is in
immediate contact with strata of water, slightly salt,
and which, following the order of superposition corre-
sponding to their temperature of maximum density
(390<1 Fahr.), are of an icy coldness. Exact observations
by Lenz and du Petit-Thouars have shown that in the
midst of the tropics, where the surface of the ocean has
a temperature of about 80°, water of 36^-° Fahr. may
be drawn up from depths of seven or eight hundred
fathoms ; a phenomenon which manifests the existence
of submarine currents from the polar regions. The
consequences of this sub-oceanic constant cooling pro-
cess of by far the larger portion of the earth's crust,
deserve a greater measure of attention than they have
hitherto received. Rocks and islands of small circum-
ference rising like cones from the bottom of the sea
above the surface of the waters, and narrow isthmuses,
such as Panama and Darien, washed by wide oceans,
must present a different internal distribution of heat, to
that existing in masses of similar form and density in
the interior of continents. In a lofty mountain-island,
the submarine portion of its vertical height is in contact
with a fluid of which the temperature increases from
below upwards ; while, as the rocky strata emerge from
the sea, they come in contact, under the influence both
of the sun's direct rays and of free outward radiation,
with a gaseous fluid in which the temperature decreases
with increasing height. Similar thermal relations, of
opposite decrease and increase in the vertical line.

o 2

196 KEACTION OF THE INTERIOR OF THE EARTH

are found in the narrow tract (Ust-Urt) intermediate
between two great inland seas, the Caspian and the
Aral. In order, at some future time, to elucidate fully
phenomena so complicated, it will be needful to employ
exclusively methods which lead directly to the know-
ledge of the internal temperature, such as deep borings,
and not to trust merely to observations of the tempera-
ture of springs, or of the atmosphere in caves ; for these
give results as insecure as does the temperature of
the atmosphere in the galleries and chambers of mines.
In comparing a low flat land with mountain-ridges
rising abruptly many thousand feet, or with high table-
lands, the law of increasing and decreasing temperature
does not depend simply on the relative heights of the
points compared. If we should compute, according to a
determinate scale, the variation of temperature for a
given number of feet taken in proceeding upwards from
the plain to the summit, and in a line downwards from
the summit to a stratum in the interior of the mass of
the mountain on a level with the surface of the plain,
we should find, in the one case, the summit much too
cold, and in the other, the interior stratum much too
warm. The distribution of heat in a mountain eleva-
tion (an undulation of the earth's surface) is dependent,
as has been already remarked, on the form and mass of
the mountain, — the conducting power of the rocks of
which it consists, — the heat received from the sun's rays
on the one hand and lost on the other by radiation, both
in measure varying with the clear or clouded character
of the atmosphere,— and on the contact and varying
play of ascending and descending currents of air:
granting these suppositions, we might reasonably expect

ON ITS EXTERIOK. THERMAL SPRINGS. 197

that, with very moderate differences of elevation, such
as four or five thousand feet for example, mountain-
springs would frequently be found, whose temperature
should be from 70° to 90° Fahr. above the mean tempera-
ture of the place ; and lastly, what might we not expect
would be the case at the foot of mountains in the torrid
zone, which, at a height of nearly 15,000 feet, are still
free from perpetual snow, and which often show no vol-
canic rocks, but only gneiss and mica-slate ? (268) The
great mathematician Fourier, the year before his death,
(at my solicitation, and stimulated by the consideration
of the topography of the eruption of Jorullo in a plain
where, for many hundred square leagues around, no
traces of unusual terrestrial heat had been discovered,)
engaged in theoretical investigations into the question of
the manner in which, in cases of upheavals and altered
configuration of the surface, the isothermal planes adjust
themselves afresh, in accordance with the new form of
the ground. The lateral radiation of strata, which are
situated at the same level but unequally covered, has
greater influence in this matter, than has the inclination
of the planes of separation of the rock where stratifica-
tion is discernible.

I have already remarked elsewhere, (269) how the hot
springs in the vicinity of ancient Carthage, probably the
thermal waters of Pertusa (aquae calidse of Hammam-
el-Enf), led the martyr, Bishop Patricius, to a correct
view as to the cause of the higher or lower temperature
of the gushing fountains. When the pro-consul Julius
sought mockingly to perplex the accused bishop by the
question, " Quo auctore fervens hsec aqua tantum ebul-
liat ? " Patricius, in reply, unfolded his theory of central

198 REACTION OF THE INTERIOR OF THE EARTH

heat, " which occasions the fiery eruptions of Etna and
Vesuvius," and " renders springs so much the warmer as
their origin is deeper." He considered the Pyriphlege-
thon of Plato to be the hell of sinners ; and further, as if
he would have recalled one of the cold hells of the
Buddhists, supposed, somewhat unscientifically, for the
" nunquam finiendum supplicium impiorum," notwith-
standing the depth, an " aqua gelidissima concrescens in
glaciem."

Among hot springs, those which attain a temperature
of 194° Fahr. (approaching the boiling point of water,
212°) are much more rare than, from inexact observa-
tions, is commonly believed, and least of all are they to be
looked for in the neighbourhood of still active volcanoes.
In my American journey, I was able to examine two of
the most Important of such springs, both within the
tropics. In Mexico, not far from the rich silver mines
of G-uanaxuato, in 21° north latitude, at an elevation of
fully 6400 feet above the level of the sea, (27°) the
Aguas de Comangillas gush forth from a mountain of
basalt and basaltic breccia. I found their temperature,
in September 1803, 205°-5 Fahr. This basaltic mass
has been broken through by a dyke of columnar por-
phyry, which itself rests on a white quartzose syenite.
Higher up, but not far from this almost boiling spring,
at los Joares, north of Santa Eosa de la Sierra, snow
falls from December to April at an elevation of only
8700 feet ; and throughout the year the natives prepare
ice, by means of radiation, in artificial basins. In the
route from Nueva Valencia in the Valles de Aragua to
the harbour of Portocabello (in about 10 \° north la-
titude), on the northern declivity of the Venezuelan

ON ITS EXTEEIOK. THERMAL SPRINGS. 199

coast chain, I saw the aguas calientes de las Trincheras
gush forth from a stratified granite which does not pass
into gneiss. I found the temperature of this spring, in
February 1800, 194°-5,(271) while the Banosde Mariara
in the Yalles de Aragua, which are in gneiss rock, showed
138°*7. Three and twenty years later, again in the
month of February, Boussingault and Eivero (272) found
at the Mariara waters 1470<2, and at the Trincheras de
Portocabello, but little above the Caribbean Sea, in one
basin 198°, and in the other 206°'6. It would appear,
therefore, that in the short interval of little more than
twenty years, an unequal change of temperature had
occurred in these springs, the increase in the tempe-
rature of the Mariara waters being 8°*5, and in that of
the Trincheras 12°*1. Boussingault rightly called atten-
tion to the fact that the dreadful earthquake, which over-
threw the town of Caracas on the 26th of March, 1812,
took place in this very interval It is true that
at the surface of the earth the commotion was less
in the neighbourhood of the lake of Tacarigua (in
Nueva Valencia); but in the interior of the earth,
where elastic vapours act in clefts and fissures, may not
a movement propagating itself so widely, and with so
much violence, easily modify the network of fissures,
and open deeper channels for the access of heat ? The
hot waters of the Trincheras rising out of a granite for-
mation are almost pure, containing only traces of silica,
sulphuretted hydrogen and nitrogen. After forming
several very picturesque cascades, bordered by luxu-
riant vegetation, they become a river, the Eio de
Aguas calientes, which as it approaches the sea is full
of large crocodiles, the still high, though already

200 REACTION OF THE INTERIOR OF THE EARTH

greatly diminished, temperature suiting those crea-
tures extremely well. In the extreme north of India
(lat. 30° 52'), the very hot spring of Jumnotri rises
out of granite, at an elevation of 10,850 feet above
the sea-level, with a temperature of 194° Fahr., being
nearly the boiling point of water under the atmospheric
pressure corresponding to that altitude. (273)

Among intermitting hot springs, the boiling fountains
of Iceland, and among these especially the Great Greysir
and the Strokr, have deservedly acquired the most
celebrity. In both, according to the excellent recent
researches of Bunsen, Sartorius von Waltershausen, and
Descloiseaux, the temperature in the jets of water de-
creases from below upward with remarkable rapidity.
The Greysir has a truncated cone, about 26 or 30 feet high,
formed of horizontal strata of silicious concretion. In
this cone there is a shallow depression about 55 or 56
feet broad, in the middle of which a funnel about 1 8 feet
in diameter descends to a depth of 75 feet. The tempe-
rature of the water, which always fills the basin, is nearly
180° Fahr. At very regular intervals of about 80 or 90
minutes, subterranean thunder announces the beginning
of the eruption. Columns of water nine or ten feet dia-
meter, of which about three large ones follow each other
in immediate succession, shoot upwards to the height of
more than a hundred, sometimes to even nearly a hun-
dred and fifty feet. The temperature of the water which
rises in the funnel has been measured at a depth
of 72-i- feet, and found, a short time before the eruption,
260°-6 ; during the eruption, 255°'6 ; and immediately
after it 251°*6 At the surface of the basin it was
only from 183° to 185°. The Strokr, which is also

ON ITS EXTERIOR. THERMAL SPRINGS. 201

situated at the foot of the Byarnefell, has less water
than the Great Greysir. The siliceous margin of its
basin is only a few inches high and wide. Its eruptions
are more frequent than those of the Greysir, but do not
announce themselves by subterranean noise. In the
Strokr the temperature at the time of eruption is from
235°-4 to 239° at a depth of 42£ feet, and almost
212° at the surface. The eruptions of the intermitting
boiling fountains, and the small variations in the type of
their phsenomena, are quite independent of the eruptions
of Hecla, and were in no ways interrupted by those
which took place in 1845 and 1846.(274) Bunsen, with
his peculiar sagacity both in observation and discussion,
has refuted the formerly received hypotheses respecting
the periodicity of the Greysir eruptions (subterranean
hollows filled as steam-boilers alternately with vapours
and with water). According to his view, the eruptions
take place because a part of a column of water, which at
some lower depth has gained a high degree of tempera-
ture under great pressure by accumulated vapours, is
pushed upwards, and thus comes under a pressure in
which a portion of the water is converted into steam.
Thus, " the Geysirs are natural collectors of steam-
power."

Among hot springs, some few are almost absolutely
pure, others contain solutions of as many as from 8 to
12 solid or gaseous substances. To the first class belong
the Luxueil, Pfeffers, and Gastein waters; of which the
nature of the efficacy appears so perplexing on account
of their purity. (275) As all springs are principally fed
by meteoric water, (water which has fallen in rain, hail,
or snow,) they contain nitrogen ; as Boussingault has

202 REACTION OF THE INTERIOR OF THE EARTH

shown to be the case in the very pure spring which flows
from granite rock in the Trincheras de Portocabello, (276)
and Bunsen (277) in the Cornelius spring at Aix-la-Cha-
pelle, and in the Geysirs of Iceland. The organic
matter which is held in solution in several springs is also
nitrogenous, and even sometimes bituminous. Formerly,
before it was known by Gay-Lussac's and my own expe-
riments, that rain and snow-water contain (the first 10
and the last at least 8 per cent.) more oxygen than the
atmosphere, it was thought very surprising that it should
be possible to obtain from the springs of Nocera, in the
Apennines, mixture of gases in which there was an
abundance of oxygen. The analyses of Gay-Lussac,
made while we resided near this mountain-spring,
showed that it contained no more than the amount of
oxygen which it could derive from rain or snow. (278) If
we are disposed to wonder at the silicious deposits which
in the Geysirs appear more like building materials, out
of which nature has put together forms resembling arti-
ficial constructions, we may be reminded that silica is
also contained in many cold springs which have in them
only a very small portion of carbonic acid.

Saline springs and exhalations of carbonic-acid gas,
which were long ascribed to beds of coal and lignite,
appear to be due wholly to processes of deep-seated vol-
canic activity; an activity, not manifesting itself solely
where volcanic rocks testify the existence of ancient
local igneous eruptions, but everywhere diffused. Car-
bonic-acid exhalations are, indeed, the phenomena which
in extinct volcanoes longest survive plutonic catastrophes :
they succeed to the stage of solfatara activity ; but at the
same time waters of the most different temperatures

ON ITS EXTERIOR. THERMAL SPRINGS. 203

issue, strongly impregnated with carbonic acid, alike
from granite, gneiss, and the older and newer sedi-
mentary rocks. Springs containing acids become im-
pregnated with carbonates, particularly with carbonate
of soda, wherever waters having carbonic acid act on
rocks which contain alkaline silicates. (279)

In the north of Germany, at many of the carbonic-
acid springs of water and gas, the still subsisting dislo-
cation of the strata and the character of the valleys in
which the springs break forth, for the most part " annular
valleys " (as at Pyrmont, Driburg, &c.), are particularly
striking. Friedrich Hoffmann and Buckland almost si-
multaneously gave to such depressions the very character-
istic name of "valleys of elevation " (Erhebungs-Thaler).

In the springs termed sulphur-waters, the sulphur is
by no means found always in the same combinations.
In many which contain no carbonate of soda, sulphu-
retted hydrogen is probably held in solution ; in others,
for example in the sulphur-waters of Aix-la-Chapelle
(in the Kaiser, Cornelius, Rose, and Quirinus springs),
the gases obtained by boiling with the air excluded, do
not contain, according to exact experiments by Bunsen
and Liebig, any sulphuretted hydrogen; and even in
the bubbles of gas which rise spontaneously from the
springs themselves, it is only in the Kaiser spring that
0-31 per cent, of sulphuretted hydrogen is obtained.(280)

A thermal spring, giving birth to an entire river of
water impregnated with sulphurous acid, the Vinegar
Eiver (Rio Vinagre), called by the natives Pusambio, is
a remarkable phsenomenon, which I was the first to
make known. The Rio Vinagre rises, at a height of
about 10,600 feet, on the north-western declivity of the

204 REACTION OF THE INTERIOR OF THE EARTH

volcano of Purace, at the foot of which the town of
Popayan stands. It forms three picturesque cascades,(281)
of one of which, which falls over a precipice more than
300 feet high, I have given a drawing. From the point
where this small stream falls into the Cauca, to the
mouths of the Pindamon and Palace 10 or 12 miles
lower down, no fish are found in that large river, much
to the annoyance of the inhabitants of Popayan, who are
strict observers of ecclesiastical fasts. The waters of the
Pusambio contain, according to Boussingault's later-made
analysis, a large quantity of hydrosulphuric acid and
carbonic acid gases as well as some sulphate of soda. Near
the source Boussingault found the temperature 163°. The
upper portion of the course of the Pusambio is subter-
ranean. In the Paramo de Euiz, on the side of the
volcano of the same, name, at the sources of the Eio
Gruali, at an elevation of 12,150 feet, Degenhardt (a
geologist from Clausthal in the Hartz, whose early death
is a loss to geological science,) discovered in 1846 a hot
spring in the waters of which Boussingault found three
times as much sulphuric acid as in the Eio Vinagre.

The persistent identity in temperature and chemical
composition of much the greater number of thermal
springs as far back as can be traced by secure observa-
tions, is a more remarkable fact than the variations
which have been made out in a few cases. (282) Thermal
springs, whose waters during their long and intricate
course take up from the rocks with which they come in
contact a variety of ingredients, which they often con-
duct to parts where the strata are wholly wanting
in such components, exert also another and entirely dif-
ferent kind of action, — an action at once metamorphic

ON ITS EXTERIOR. THERMAL SPRINGS. 205

and formative. In this point of view they are of great
geological importance. Senarmont has shown with ad-
mirable sagacity the manner in which it is highly proba-
ble that many veins, ancient channels of thermal waters,
have been rilled from below upwards by deposits of the
elements which had been held in solution. By changes
of pressure and temperature, by internal electro-che-
mical processes, and by the specific attraction of the
side walls, lamellar divisions have taken place and con-
cretions have been formed. Drusic cavities and porous
amygdaloids seem to have been formed partly in this
manner. When the deposit in veins has taken place in
parallel zones, these zones correspond to each other as
respects their constitution, for the most part symme-
trically from the sides.

Senarmont's chemical ingenuity succeeded in obtaining
artificially a considerable number of minerals by quite
analogous, synthetic processes. (283) I hope that a scienti-
fically gifted observer, nearly connected with myself,
will shortly publish a new and important investigation
respecting the temperatures of springs, in which, by in-
duction from a long series of observations, he will treat
the intricate phenomena of disturbing effects with great
generality. Eduard Hallmann, in the thermoinetric
measurements obtained by him during the years from
1845 to 1853 in Germany (on the Rhine), and in Italy (in
the country round Rome, in the Alban hills, and in
the Apennines), distinguished between, 1st, pure meteor-
ological springs, whose mean temperature is not raised
by the internal temperature of the earth ; — 2nd, mete-
orolo-geological springs; which being independent of
the distribution of rain, and warmer than the atmo-

206 REACTION OF THE INTERIOR OF THE EARTH

sphere, undergo only such changes of temperature as
are imparted to them by the ground through which
they issue forth ; — 3rd, abnormally cold springs, in
which the low temperature is brought down from great
heights. (284) The greater the advance which has been
made of late years, by the happy application of chemistry
to geology, towards a knowledge of the mode of forma-
tion, and of the metamorphic alteration, of rocks, the
greater has been the importance attaching to the consi-
deration of those spring waters which circulate through
the crust of the earth, charged with different salts and
gases, and which when they break forth at the surface
as thermal waters have already completed the greater
part of their formative, metamorphic, or destructive
functions.

c. Springs of Vapour and Gas, Salses, Mud-volca-
noes, Naphtha-flames.

(Enlargement of the Picture of Nature in Cosmos,
Vol. I. pp.211— 213.)

In the general picture of nature in the first volume of
this work I have shown, by examples which have been
well made out, although they have not been sufficiently
attended to, that " Salses " in the different stages through
which they pass, from eruptions accompanied by flames
to mud emitted quietly, form, as it were, an interme-
diate link between hot springs and volcanoes proper ;
which latter throw up molten earths, in the shape
of disconnected scoriae, or, as newly formed and often
variously superposed rocks. Like all transitions and inter-

ON ITS EXTERIOR. SPRINGS OF YAPOUR AND' GAS. 207

mediate links in inorganic and organic nature, salses and
mud-volcanoes are well-deserving of a serious attention,
which the older geologists, for want of special knowledge
of the facts, were not so well able to bestow on them.

Salses and naphtha-springs are found, in some cases,
in small detached groups; as the Macalubi in Sicily,
near Girgenti (referred to so early as by Solinus), those
near Pietra Mala, Barigazzo and at the Monte Zibio,
not far from Sassuolo in the north of Italy, and at Tur-
baco in South America ; — and in other cases (which
are the more important and instructive) arranged as
it were in long narrow lines. The mud-volcanoes of
Taman on the north-west of the chain of the Caucasus,
and the naphtha-springs and burning naphtha of Baku
and of the Caspian peninsula of Apscheron on the
south-east, have long been known (285) as the extreme
links belonging to that great mountain-chain ; but
the magnitude and connection of these phsenomena
were first shown by Abich, who is profoundly ac-
quainted with that portion of western Asia. Accord-
ing to his views, the Caucasian mud-volcanoes and
founts of burning naphtha are recognisably connected
with certain lines, which are in unmistakable relation
to the "axes of elevation" and "directions of dis-
location" of the strata. The larger portion of the
south-eastern part of the Caucasus is occupied by gene-
tically connected mud-volcanoes, naphtha exhalations,
and saline springs in a triangle with two equal sides,
having for its base the shore of the Caspian from Bala-
chani (north of Baku) to one of the mouths of the
Kur (Araxes) near the hot springs of Sallian, and for
its apex Schagdagh in the high valley of Kinalughi.

208 REACTION OF THE INTERIOR OF THE EARTH

Not far from the village of Kinalughi, at an elevation of
nearly 8350 feet above the Caspian and at the junction of
a dolomite and a slate formation, there break forth " the
perpetual fires of Schagdagh," which are never extin-
guished by the weather. The axis of this triangle cor-
responds to the direction which appears to be constantly
followed by the earthquakes, which are so often experi-
enced in Schamacha, on the banks of the Pyrsagat. If
this north-western direction is prolonged, it encounters
the hot sulphurous springs of Akti, and then becomes
the line of the principal crest of the Caucasus, where it
rises to the Kasbegh and bounds the western part of
Daghestan. The salses of the low country, often regu-
larly arranged in line, become gradually more frequent
towards the shore of the Caspian between Sallian,
the mouth of the Pyrsagat (near the island of Swinoi),
and the peninsula of Apscheron. They show traces
of former repeated eruptions of mud, and bear on their
summits small cones, quite similar in shape to the hor-
nitos of Jorullo in Mexico, from which there exhales an
inflammable gas, which indeed often ignites spontane-
ously. Considerable outbreaks of flames were particu-
larly frequent between 1844 and 1849 on the Oudplidagh,
Nahalath, and Turandagh. Close by the mouth of the
Pyrsagat, at the mud-volcano of Toprachali, there are
found (as proofs of a greatly increased intensity of sub-
terranean heat) " black pieces of marl, which at first
sight might be mistaken for dense basalt and exceed-
ingly fine-grained dolerite." At other parts of the
peninsula of Apscheron, Lenz found fragments which
looked like erupted scoriae, and at the time of the
great burst of flames at Baklichli on the 7th of Fe-
bruary 1839, small hollow balls, like what are called

ON ITS EXTERIOR. SPRINGS OF VAPOUR AND GAS. 209

ashes in regular volcanoes, were carried by the wind to a
considerable distance.(286)

At the north-western extreme, towards the Cimmerian
Bosphorus, are the mud-volcanoes of the Taman penin-
sula, which group themselves with those of Aklanisowka
and Jenikale near Kertch. At one of the salses of
Taman there was an eruption of mud and gas on the
27th of February 1793, in which, after much subterra-
nean noise, a column of flame several hundred feet
high, rose into the air half-veiled in black smoke. (May
not the supposed smoke have been a dense cloud
of steam ?) It is a remarkable circumstance, and a very
instructive one in regard to the nature of the volcancitos
of Turbaco, that the gas tested at Taman in 1811,
by Friedrich Parrot and Engelhardt, was not inflam-
mable ; whereas at the same place, 23 years later, gas
received by Grobel in a glass tube burnt, out of the
tube, with a bluish flame, like the exhalations from
all the salses in the south-eastern part of the Caucasus ;
but when subjected to precise analysis was found
to contain out of 100 parts 92*8 of carburetted hydro-
gen gas and 5 parts of carbonic oxide gas. (287)

A phenomenon assuredly allied in respect to origin to
those of which we are treating, although differing in the
substances produced, is that of the eruptions of hot bo-
racic vapours in the Maremma of Tuscany, known
under the names of " lagoni," " fumerole," " soffioni,"
and also " volcani," at Possara, Castel Novo and Monte
Cerboli. The vapours have average temperatures vary-
ing from 204°*8 to 212°, and at a few points amounting
even to 347°. They rise in some cases directly from
clefts in the rocks, and in others in pits, in which they

VOL. iv. P

210 REACTION OF THE INTERIOR OF THE EARTH

throw up small cones formed of liquid mud. They are
seen to rise in whitish whirls of vapour which disperse
themselves in the air. The boracic acids, which these
aqueous vapours bring up with them from the bosom of
the earth, cannot be secured when the vapours of the
soffioni are collected in vessels placed above the orifices ;
for they are so volatile that they escape into the atmo-
sphere. The acid is only successfully obtained in the
skilfully arranged establishments belonging to Count
Larderel, where the orifices of the soffioni are covered by
artificially formed reservoirs of water, which absorb the
vapours. (m) According to Payen's excellent analysis,
the gaseous exhalations contain 0'57 carbonic acid, 0'35
nitrogen, only 0-07 oxygen, and 0*001 sulphuric acid.
Where the boracic vapours penetrate the fissures of the
rock they deposit sulphur. According to Sir Eoderick
Murchison, the rock is partly cretaceous, and partly an
eocene formation containing nummulites; a "macigno,"
broken through by upheaved serpentine, which shows
itself at the surface in the vicinity at Monte Eotondo.(289)
May we not suppose, it is asked by Bischof, that here,
and in the crater of Vulcano, hot steam (aqueous
vapour) acts at a great depth on minerals containing
boracic acid, or on rocks containing datholite, arinite,
and tourmaline, and decomposes them ? (29°)

The system of soffioni in Iceland exceeds, both in
variety and magnitude, anything with which we are
acquainted on the continent. In the Fumarole-Field of
Krisuvek and Eeykialidh true fountains of mud break
forth out of greyish blue clay from small basins with
crater-shaped margins. (291) Here also definite di-
rections can be pointed out, along which fissures
marked by phenomena of this class can be traced. (2/)2)

ON ITS EXTERIOR. SPRINGS OF VAPOUR AND GAS. 211

There is no part of the earth's surface where hot
springs, salses, and eruptions of gas are found, for
which we now possess such excellent and detailed che-
mical examinations as those for Iceland, which we owe
to the sagacity and persevering labours of Bunsen.
Nowhere, so far as we know, are such varied actions
of chemical decomposition, formation, and transform-
ation to be observed over so great a tract of country,
and probably in such near proximity to the surface.

Passing from Iceland to the American continent, we
find in the state of New York, in the district of Fre-
donia, not far from Lake Erie, in a basin of strata of
Devonian sandstone, an immense number of springs
of inflammable gas, (springs of carburetted hydro-
gen), breaking forth from fissures, and sometimes used
for gas-lights ; other springs of combustible gas, as at
Eushville, take the shape of mud-cones ; others, as in
the valley of the Ohio, in Virginia, and at the Ken-
tucky river, contain at the same time common salt, and
in such case are connected with weak naphtha-springs.
Passing to the south of the Caribbean Sea, on the north
coast of South America, ten miles S.S.E. of the port of
Cartagena de Indias, near the pleasant village of Tur-
baco, a remarkable group of salses, or mud-volcanoes,
presents phenomena which I was the first to describe.
At this place, where a glorious view of the colossal
snow-covered mountains (Sierras Nevadas) of Santa
Marta is enjoyed, there rise on a desert space, in the
midst of the primeval forest, 18 or 20 " volcancitos."
The larger of these cones of dark grey are from 19 to
23|- feet high, and fully 85 feet in diameter at their
base. At the top of each cone there is a circular

p 2

212 REACTION OF THE INTERIOR OF THE EARTH

opening of from 21^- to 30 inches diameter, surrounded
by a little mud wall. The gas rises with great vehe-
mence, as at Taman, in bubbles, each of which, accord-
ing to my measurements made in graduated glass vessels,
contains from 10 to 12 (French) cubic inches. The
upper part of the funnel is filled with water resting on
a thick mud covering. Adjacent cones do not have
simultaneous eruptions, but in each single cone there
may be remarked a certain degree of regularity in the
times of eruption. Bonpland and I counted pretty re-
gularly five eruptions every two minutes. When one
stoops over the little crater opening, one can perceive,
generally about twenty seconds before each eruption, a
hollow noise in the interior of the earth deep below the
base of the cone. In the gas which rose, and which was
on two occasions collected with great care, a lighted
very thin wax taper was instantly extinguished, as was
also a burning splinter of Bombax Ceiba. The gas
could not be made to ignite. It did not render lime-
water turbid ; no absorption took place. When tested
by nitric oxide gas for oxygen, it showed no trace of the
latter ; in another experiment, in which the gas of the
volcancitos was kept for many hours in a small bell-glass
turned down in water, it showed rather more than one
per cent, of oxygen, which was probably derived from
the water, and had become accidentally mixed with it.

According to these results of analysis, I then stated
that I considered (not altogether incorrectly) the gas
of the volcancitos of Turbaco to be nitrogen, which
might be mixed with a small quantity of hydrogen. * I
expressed at the same time in my journal regret, that
in the then state of chemical knowledge (in April 1801)

ON ITS EXTERIOR. SPRINGS OF VAPOUR AND GAS. 213

there was no known way of determining, in a mixture of
nitrogen and hydrogen, the relative proportions of each.
A method of doing this, so that an admixture of three
thousandths of hydrogen could be detected, was dis-
covered four years afterwards by Gray-Lussac and my-
self. (293) In the half century which has elapsed since
my visit to Turbaco and my astronomical survey of the
Magdalena Eiver, no traveller examined the above-de-
scribed small mud- volcanoes scientifically ; until, at the
end of December 1850, my friend Joaquin Acosta, (294)
who was acquainted with modern geology and chemistry,
made the important observation that, — " at that time
the cones exhaled a bituminous odour; that some petro-
leum floated on the surface of the water in the orifices ;
and that the gas which rose from each of the volcancitos
was inflammable." Of all this there was no trace in my
time. Acosta asks whether these circumstances indicate
a real alteration produced in the phenomena by subter-
ranean processes, or whether the discrepancy arises
merely from error in the earlier experiments ? I should
have freely admitted the latter to have been a probable
explanation, if I had not preserved the leaf of the journal
in which I noted the experiments in detail on the
same morning on which they were made. (295) I find
nothing therein which could now give me any doubts
respecting them ; and the experience above alluded to,
i. e. that (according to Parrot's account) " the gas of
the mud-volcanoes of the peninsula of Taman had, in
1811, the property of preventing combustion; that a
glimmering splinter was extinguished in it ; and that the
bubbles which rose a foot thick, could not be made to
ignite in bursting," — whereas at the same place, Gobel,

214 REACTION OF THE INTERIOR OF THE EARTH

in 1834, saw the gas easily kindled and burning with a
clear blue flame, — leads me to believe that, in dif-
ferent stages, the exhalations undergo chemical altera-
tions. Mitscherlich, at my request, has recently de-
termined the limits of inflammability in artificially
prepared mixtures of nitrogen and hydrogen gases.
The result was that mixtures of one part hydrogen and
three parts nitrogen would not only be ignited by a light
being brought near to them, but also would continue to
burn : when the proportion of nitrogen was augmented
so that the mixture consisted of one part hydrogen and
three and a half parts nitrogen, it would still ignite,
but would not continue to burn. It was not until the
mixture consisted of one part hydrogen and four parts
nitrogen that no ignition at all took place. The gaseous
exhalations, which, on account of their easy inflamma-
bility and the colour of their light, are usually called ex-
halations of pure and carburetted hydrogen, need there-
fore only in fact have as much as one third part of that
gas in their composition. In the more rare mixtures of
carbonic acid and hydrogen, on account of the capacity
for heat of the former, the limit of capability of igni-
tion is modified. Acosta is right in throwing out the
question, "whether a tradition subsisting among the
natives of Turbaco, descendants of the Indios de
Taruaco, according to which the volcancitos were once
all burning, and were all changed from 'volcanos de
fuego' into f volcanos de agua' by the adjurations of a
pious monk, and being sprinkled by him with holy
water, (296) may not have related to a former condition
which is now returning ? " Analogous examples are pre-
sented by, at one time the eruption of great flames, and

ON ITS EXTERIOR. SPRINGS OF VAPOUR AND GAS. 215

subsequently very peaceful mud-volcanoes, in the region
of the old continent already referred to. (Taman, 1793 ;
on the Caspian at Jokmali, 1827; at Baklichli, 1839 ;
and at Kuschtschy in the Caucasus, 1846.)

The apparently small phenomenon of the salses of
Turbaco has gained in geological interest by the great
outbursts of flame and changes of surface, which in 1839
extended between thirty and forty miles N.N.E. of
Carthagena to the port of Sabanilla, not far from the
mouth of the great Magdalena River, and for the know-
ledge of which we are also indebted to the late lamented
Artillery-Colonel, Acosta. The proper central point of
this latter phsenomenon appears to have been situated
in the narrow peninsula called Cape Gralera Zamba, which
runs out six or eight miles into the sea. In the middle
of this tongue of land there stood a conical hill having
a crater-like opening, from which smoke (vapours and
gases) sometimes rose with such vehemence, that boards
and large pieces of wood, which had been thrown into
it, were expelled; and hurled to a great distance. In the
year 1839 this cone disappeared, its disappearance being
accompanied by a considerable outbreak of fire ; and the
whole peninsula of Gralera Zamba became an island,
separated from the mainland by a channel more than
thirty feet deep. The surface of the sea remained tran-
quilly in this state until, on the 7th of October 1848,
without any earthquake being felt in the country round,
there appeared a second great outburst of flames, (297)
which lasted several days and were visible to distances
of between forty and fifty miles. The salse emitted only
gases, not solid or liquid substances. When the flames
had ceased, the sea-bottom was found to have been

216 REACTION OF THE INTERIOR OP THE EARTH

upheaved in the form of a small sandy island, which,
however, disappeared again soon afterwards. More than
fifty " volcancitos " (cones similar to those of Turbaco)
now surround (in some cases at a distance of sixteen or
twenty miles) the submarine gas-volcano of the Galera
Zamba. This may well be regarded as, in geological
respects, the principal seat of the volcanic activity
which throughout the low-lying district of Turbaco,
and extending over the delta of the Eio Grande de la
Magdalena, strives to place itself in contact with the
atmosphere.

The similarity of the phenomena shown, in the dif-
ferent stages of their activity, by the salses, mud-vol-
canoes, and gas-springs on the Italian peninsula, in the
Caucasus, and in South America, is displayed over vast
tracts of country in the Chinese empire. There the art
of man has, since very early times, utilised these great
natural supplies; and for this object had arrived long
since at the very ingenious invention with which Euro-
peans only became acquainted at a far later period, —
that of " Chinese rope-boring." These borings, several
thousand feet deep, were effected by the simplest ap-
plication of man's bodily power, or rather of his weight.
I have elsewhere (298) treated in detail of this invention,
as well as of the " fire-springs " (Ho-tsing), and of the
fire-mountains (Ho-schan), of Eastern Asia. Men bore
for water, salt, and combustible gas, from the south^
west provinces of Yun-nan, Kuang-si, and Szu-tchuan
on the borders of Thibet, to the northern province of
Schan-si. The gas burns with a reddish flame and often
diffuses a bituminous smell : it is conveyed to a distance,
sometimes through pipes of bamboo, sometimes in port-

ON ITS EXTERIOR. SPRINGS OF VAPOURS AND GASES. 217

able tubes also of bamboo, to be used in salt-works, in
warming houses, or in lighting streets. In some rare
instances the supply has suddenly become exhausted, or
has been stopped by earthquakes. Thus it is known
that a celebrated Ho-tsing, situated south-west of the
town of Khiung-tscheu (N. lat. 50° 27', long. 103° 27' E.),
a saline spring which burnt with noise, became extinct in
the 13th century, after having supplied lights for the
country round for 1100 years. In the province of
Schan-si, which is very rich in stone coal, some coal-
beds are on fire. The "fire mountains" (Ho-schan),
are found over a large part of China. The flames often
rise, at great heights above the sea (for example, in the
rocks of the Py-kia-schan, at the foot of a mountain
covered with perpetual snow in lat. 31° 40'), from long,
open, inaccessible fissures or clefts : a phenomenon which
recalls the " perpetual fires of Schagdagh " in the Cau-
casus.

In the island of Java, in the province of Samarang
about 12 miles from the north coast, there are salses
which resemble those of Turbaco and Gralera Zamba.
Hillocks, from 27 to 33 feet high (which undergo fre-
quent changes), cast up mud, salt water, and a remark-
able mixture of hydrogen and carbonic acid gases, (299) a
phenomenon which must not be confounded with the
great and devastating torrents of mud, poured forth on
the rare occurrence of eruptions of the colossal true
volcanoes of Java (Grunung Kelut and Grunung Idjen).
Some mephitic grottoes, or springs of carbonic acid gas,
in the island of Java, are still very celebrated, in great
measure from the exaggerated descriptions of some tra-
vellers, as well as from a connection, suspected by Sykes

218 REACTION OF THE INTERIOR OF THE EARTH

and by London, with the fabulous stories of the Upas or
poison-tree. The most remarkable of these, situated in
the mountains of Dieng, near Batur, has been scientifically
described by Junghuhn; it is called Pakaraman,the Island
Valley of Death. It is a fallen-in, funnel-shaped hol-
low, situated on the side of a mountain, in which the
height of the unrespirable stratum of carbonic acid gas
varies very much at different seasons. Skeletons of
wild swine, tigers, and birds are often found there. (30°)
The poison-tree, pohon or piihn, the upas of the Malays
(the Antaris toxicaria of the traveller Leschenault de la
Tour), and its harmless exhalations, have nothing what-
soever to do with these fatal effects. C301)

I will conclude this section on salses and on gas and
vapour springs with a description of a case which may
be interesting to geologists on account of the kind of
rock from which hot sulphureous vapours are developed.
When making the delightful, but somewhat arduous,
passage across the central Cordillera of Quindiu, from
the valley of the Rio Magdalena to the Cauca valley (it
took me 14 or 15 days' foot travelling, sleeping always
in the open air, to pass over the crest, 11,497 feet above
the sea), I visited the Azufral, situated, at a height of
6810 feet, on the west of the station of El Moral. In a
rather dark-coloured mica-slate, which is superposed
on a garnet-bearing gneiss, and, together with it, sur-
rounds the high granite cupolas of La Ceja and La
Garita del Paramo, I saw in the narrow valley (Que-
brada del Azufral) warm sulphureous vapours issue
from the clefts in the rock. As the vapours are mixed
with sulphuretted hydrogen and much carbonic acid gas,
a stupefying sensation of giddiness, felt on stooping down

ON ITS EXTEEIOR. VOLCANOES. 219

to measure their temperature, warns the observer not to
linger. The temperature of the sulphureous vapours
was 117°*7, that of the air 69°, and that of the " sulphur
rivulet," which, perhaps, in its upper portion may be
cooled by the snow-water from the volcano of Tolima,
84°-6. The mica-slate which contains pyrites is inter-
spersed with sulphur in pieces. The sulphur prepared
for sale is obtained chiefly from an ochrous-yellow clay
mixed with native sulphur and weathered mica-slate.
The labourers (who are Mestizos) suffer in their eyes
and muscles. When Boussingault visited the Azufral de
Quindiu, 30 years later, in 1831, the temperature of
the vapours (which he analysed chemically) (302) had
diminished so much that it was actually below that of
the free air, the latter being 7l°-6, while the former was
from 66°*2 to 68°. The same excellent observer saw, in
the Quebrada de Aguas calientes, the locality where
the trachyte of the adjacent volcano of Tolima breaks
through the mica. The also eruptive black trachyte of
the volcano of Tunguragua was distinctly seen by myself
covering the garnetiferous greenish mica-schist, near the
rope bridge of Penipe. As sulphur had not previously
been seen in Europe in what were formerly called
'" primitive rocks," but only in tertiary limestone, gyp-
sum, conglomerates, and in true volcanic rocks, its oc-
currence in the Azufral of Quindiu (N. lat. 4-^°) is note-
worthy, and the more so, because it is repeated south of
the equator, between Quito and Cuenca, on the northern
declivity of the Paramo del Assuay. In the Azufral of
Cerro Cuello (S. lat. 2° 13') I saw, also in the mica-slate,
at a height of 7980 feet, a very considerable bed of
quartz (303) richly interspersed with sulphur. At the

220 REACTION OF THE INTEEIOR OF THE EARTH

period of my visit pieces of sulphur were only found
from 6J to 8J inches in size; pieces of three or four
feet had been found earlier. Even a spring of naphtha
may be actually seen to rise from mica-schist at the
bottom of the sea in the Grulf of Cariaco, near Cumana.
The naphtha gives a yellow colour to the surface of the
sea for a distance of more than 1000 feet, and I found
its smell diffused so far as the interior of the peninsula
pf Araya. (304)

If we now cast a last glance on that kind of volcanic
activity which manifests itself by the emission of vapours
and gases, either with or without igneous phsenomena,
we find sometimes great affinity, and sometimes an
equally remarkable diversity, in the escaping substances,
according as the high internal temperature has exerted
its modifying influence on the mutual action either of
homogeneous or of highly diversified materials. The
substances which this low degree of volcanic activity
brings to the surface are — aqueous vapour or steam in
large quantities, chloride of sodium, sulphur, carburetted
and sulphuretted hydrogen, carbonic acid and nitrogen,
naphtha (colourless, yellow, or as brown petroleum),
boracic acid gas, and clay of the mud-volcanoes. The
great diversity of these substances, of which some, how-
ever (common salt, sulphuretted hydrogen, and petro-
leum), are almost always associated together, shows the
inappropriateness of the term " salse," derived from
Italy, where Spallanzani had the great merit of first
directing the attention of geologists to the previously dis-
regarded phenomena of this kind in the territory of Mo-
dena. The terms "gas-springs" and "vapour-springs"
are more generally expressive. While doubtless many

ON ITS EXTERIOE. VOLCANOES. 221

may be regarded in the light of " fumeroles " connected
with extinct volcanoes, more particularly as springs of
carbonic acid gas are a characteristic of one of the latest
stages of such volcanoes, others, on the other hand, as
naphtha-springs, appear to be wholly independent of
genuine lava-erupting volcanoes. They follow (as Abich
has already shown in the Caucasus) definite directions over
extensive spaces, breaking forth from fissures, on the plain,
and even in the deep basin of the Caspian, as well as on
mountain heights of nearly 8000 feet. Like volcanoes
proper, their apparently slumbering activity sometimes
suddenly displays itself by the bursting forth of columns
of flame, which spread terror afar. They show the
same successive stages in both continents, at parts of the
earth very remote from each other ; but hitherto expe-
rience has not authorised us to regard them as foreshow-
ing the rise of true volcanoes erupting lavas and scoriae.
Their activity is of another kind ; perhaps originating at
lesser depths, and dependent on other chemical pro-
cesses.

d. Volcanoes, according to their Diversity in Shape
and Activity — Action through Fissures and Maars
— Encircling Ridges round Craters of Elevation —
Volcanic, conical, and dome-shaped Mountains,
with open or unopened Summits — Diversity of
Rocks through which Volcanoes act.

(Enlargement of the Picture of Nature in Cosmos,
Vol.I. p. 213—235.)

Among the manifold kinds of manifestation of force in
the reaction of the interior of our planet against its

222 REACTION OF THE INTERIOR OF THE EARTH

outermost strata, the most powerful is that exhibited by
volcanoes proper, i.e. openings through which (besides
gases) solid substances of various kinds are brought in a
state of igneous fluidity from unknown depths to the
surface, where they issue forth either as streams of lava,
or as scorise, or in a state of the finest comminution,
when they are called "ashes." When, according to
the sense attached by ancient usage to the expressions
volcano and burning mountain, these are employed as
synonymous terms, the idea of volcanic phenomena is
connected, according to what was a very generally re-
ceived impression, with the image of an isolated conical
mountain, having at its summit a circular or oval open-
ing or crater. Such views lose, however, some of their
generality, when the observer has the opportunity of wan-
dering through volcanic regions of many thousand square
miles in extent ; as, for example, the entire central part
of the Mexican highlands between the Peak of Orizaba,
Jorullo, and the shores of the Pacific; or in Central
America ; or in the Cordilleras of New Granada and
Quito between the volcano of Purace near Popayan, that
of Pasto, and Chimborazo ; or in the Caucasus between
Kasbegk, Elbourz, and Ararat. In Lower Italy, between
the Phlegrsean Fields of the Campanian mainland, Sicily,
the Lipari and Ponza Isles, as well as in the Greek
islands, the intervening land has partly not been up-
heaved at the same time, and partly has been since
swallowed up by the sea.

In the above-mentioned large districts in America
and in the Caucasus we find erupted masses (true
trachytes, not conglomerates of trachyte), streams of
obsidian, and blocks of pumice-stone obtained by

ON ITS EXTERIOR. VOLCANOES. 223

quarrying (not decomposed rolled pumice-stone brought
by water), appearing to be quite independent of the
mountains, which are seen rising only at a considerable
distance. Why may we not suppose that, in the pro-
gressive cooling«down of the heat-radiating outer strata
of the earth, before either isolated mountains or moun-
tain chains were yet upheaved, the surface may have
been variously split up and fissured ? and why may
not these fissures have extruded igneously fluid masses
which have hardened into volcanic and other rocks
(trachytes, dolerites, melaphyres, pearlstone, obsidian,
and pumice) ? Part of these, originally in horizontal
beds (trachytes or dolerites which had burst forth in
a tenacious semi-fluid state, as from earth-springs), (305)
may, at the later upheaval of volcanic cone- and dome-
shaped mountains, have been thrown into a broken and
precipitous state, not seen in the later lavas which have
originated in burning mountains or volcanoes proper.
Thus, to recall first a well-known European example, —
in the Val del Bove on Etna (a hollow which cuts deep
into the interior of the mountain), the angle of " fall "
of the lava strata, alternating very regularly with rolled
masses, is from 25° to 30° ; whereas that of the streams
of lava which cover the surface of Etna, and which have
flowed since its upheaval as a mountain, have, on a
mean of thirty streams, according to Elie de Beau-
mont's exact determinations, only an angle of descent of
from 3° to 5°. Such differences point to the existence
of very ancient volcanic formations which had burst
forth from fissures before the elevation of the existing
volcano. Classical antiquity also presents us with an
example of a remarkable phenomenon of this class,

224 REACTION OF THE INTERIOR OF THE EARTH

which, showed itself in a wide plain in a district remote
from any active or extinct volcanoes ; in the island of
Euboea, the modern Negropont. " The violent shocks
of earthquake, which partially shook the island, did not
cease until an abyss, opened in the plain of Lelantus,
poured forth a stream of glowing mud (lava)." (30G)

If, as I have long been disposed to conjecture, we
may ascribe to a first fissuring of the deeply disturbed
earth-crust the oldest formations of eruptive rocks (often
perfectly similar in mineral composition to recent lavas),
then these fissures, as well as the craters of elevation of
later origin and already less simple, must be looked
upon only as volcanic openings through which erupted
masses have flowed, and not as volcanoes proper. The
principal character of these latter consists in a per-
manent, or at least from time to time renewed, con-
nection between the deep-seated focus or source of
igneous action, and the atmosphere. The volcano re-
quires for this purpose a particular kind of framework ;
for, as Seneca says very appositely in a letter to
Lucilius, f( ignis in ipso monte non alimentum habet,
sed viam." (307) In regard to volcanoes, therefore, the
form-giving, or shaping, activity is exerted by the
upheaval of the ground; not (as was formerly and
almost exclusively believed) in building up by suc-
cessive accumulation of scoriae and strata of lava
deposited one above another. The resistance which
the fiery-fluid masses, pressed in too great abundance
against the surface, find in what is to be the channel of
eruption, occasions the augmentation of the upheaving
force. There arises a "bubble-shaped pushing-up of
the ground," as is indicated by the regular outward

ON ITS EXTERIOR. VOLCANOES. 225

slope of the upheaved strata. A mine-like explosion,
the bursting of the central and highest portion of the
convex swelling of the ground, produces sometimes only
what Leopold von Buch has termed a " crater of eleva-
tion ;" (308) i.e. a crater-shaped, round or oval, depression
bounded by an " elevation-circus," a ring-shaped, and
in most cases partly broken down, rampart or encircling
ridge, and sometimes also at the same time — when
the structure of a permanent volcano is to be completed
— a dome-shaped or conical mount in the middle of
the crater of elevation. This inner mount is, in the
greater part of such cases, open at its summit, and on
the floor of this opening (or the floor of the crater of
the permanent volcano), there rise non-permanent hills
of erupted substances and scoriae, small and great
" cones of eruption ; " in Vesuvius these sometimes rise
much higher than the crater-walls of the cone of eleva-
tion. It is not always, however, that all the witnesses
to the first eruption — all the different parts of the old
framework or scaffolding, such as I have here described,
are preserved. In many of the greatest and most active
volcanoes the high escarpment forming the periphery of
the crater of elevation cannot be traced even in frag-
ments or ruins.

A great service has been rendered in modern times,
not only by the careful comparison of volcanoes situated
at remote distances from each other, and by the more
exact examination of their several relations of form ; but
also by the introduction of more definite expressions,
whereby significant differences in relief, as well as in the
manifestations of volcanic activity, are distinguished
apart. If one is not decidedly averse to all classification, —

VOL. iv. Q

226 REACTION OF THE INTERIOR OF THE EARTH

because,, notwithstanding the endeavour to generalise, it
still must always rest on imperfect inductions, — one may
represent to oneself the bursting forth of igneously
fluid masses and solid substances, accompanied by
vapours and gases in various manners. Passing from
the more simple to the more complex phenomena, we
will specify, first, eruptions from fissures, not giving
rise to a series of detached cones, but pouring forth
substances in a state of tenacious semi-fluidity, forming
superposed volcanic masses ; secondly, eruptions through
cones of accumulation without surrounding escarpment,
but yet pouring forth streams of lava, as was the case
for five years at the devastation of the island of Lan-
cerote in the first half of the last century; thirdly,
craters of elevation with upheaved strata without any
central cone, sending forth streams of lava only on the
outer side of the escarpment, never from the interior,
which is soon closed by falling in; fourthly, closed
dome-shaped mountains, or cones of elevation open at
the summit, either surrounded by an 'at least partially
preserved encircling ridge, as at the Peak of Tenerifie, in
Fogo, and Eocca Monfina, or else quite without circum-
vallation and without any crater of elevation, as in
Iceland, (309) in the Cordilleras of Quito, and in the
middle part of Mexico. The open cones of elevation
of this fourth class maintain a permanent more or
less active communication, in indeterminate intervals
of time, between the heated interior of the earth and
the external atmosphere. Dome-shaped trachyte and
dolerite mountains with closed summits seem, according
to what I have been able to observe, to be more
numerous than open cones, either still active or extinct,
and much more numerous than volcanoes proper. Dome-

ON ITS EXTERIOR. VOLCANOES. 227

and bell-shaped mountains, like Chimborazo, Puy de
Dome, Sarcouy, Eocca Monfina, and Vultur, give to the
landscape in which they are seen a peculiar character,
by which they contrast agreeably with the schistose
Horns or the jagged forms of calcareous rock.

The elevation of such a dome-shaped mount without
opening is indicated with great clearness in the very
graphic description in which Ovid has preserved the tra-
dition of a great volcanic event in the peninsula of
Methone. " The powerful force of winds imprisoned
in dark caverns, seeking in vain an outlet, caused the
inflated ground to swell upwards, as when a bladder or
a goat-skin is filled with air. The high swelling by
slow hardening has remained a hill." I have remarked
elsewhere how entirely this Eoman description differs
from Aristotle's account of the volcanic event at Hiera,
a newly arisen ^Eolian island : " the subterranean power-
fully urging breath" is indeed here also described as
raising a hill, but also as " afterwards breaking it up and
pouring from it a fiery rain of ashes." Here the eleva-
tion is distinctly described as preceding the igneous
outburst. (Cosmos, Vol. I. note 230.) According to
Strabo it would seem as if the dome-shaped hill of
Methana had also opened in a fiery eruption, at the
termination of which an agreeable odour diffused itself
each night. It is a striking circumstance that a sweet
smell was remarked under similar conditions at the
volcanic eruption of Santorin in the autumn of 1650,
and is mentioned in a sermon, preached soon afterwards
by a monk, which has been preserved. (31°) May not
•such an odour indicate naphtha ? The same thing is
spoken of by Kotzebue in his Russian Voyage of Dis-

Q 2

228 REACTION OF THE INTERIOR OF THE EARTH

covery, on the occasion of an igneous eruption, in 1804,
of the island-volcano of Umnack, then newly risen from
the sea, in the Aleutian Archipelago. At the great
eruption of Vesuvius on the 12th of August 1805,
which Gray-Lussac and I observed together, he perceived
from time to time a predominating bituminous smell in
the burning crater. I have here brought together these
few and hitherto little regarded facts, because they help
to confirm the close interconnection of all manifestations
of volcanic activity, linking salses and naphtha-springs
with actual volcanoes.

Encircling ridges analogous to those of craters of
elevation, show themselves in kinds of rock very dif-
ferent from trachyte, basalt, and porphyritic schist;
for example, according to Elie de Beaumont, in the
granite of the French Alps. The Oisans mass of
mountains, to which the highest summit in France, (3H)
Mont Pelvoux near Brianpon (12,905 feet), belongs,
forms a circus of thirty-two geographical miles in cir-
cumference, in the middle of which lies the little village
of La Berarde. The steep walls of the circus rise to
a height of 9600 feet. The circumvallation itself is
gneiss; all within it is granite. (312) In the Swiss and
Savoy Alps the same form is shown in several cases,
in smaller dimensions. The Grand Plateau of Mont
Blanc, on which Bravais and Martins encamped several
days, is a closed circus, or amphitheatre, 12,810 feet
high, with an almost even floor, out of which the great
summit-pyramid rises. (3I3) The same upheaving forces
produce similar forms, though modified by the compo-
sition of the rocks on which they have been exerted.
The valleys of elevation, described by Hoffmann, Buck-

ON ITS EXTERIOR. VOLCANOES. 229

land, Murchison, and Thurmann, in the sedimentary
rocks of the North of Germany, in Herefordshire, and
in the Jura Mountains of Porrentruy, are connected
with the phenomena which have been here described ;
as are also, though with a less measure of analogy, some
high plains in the Cordilleras, closed in on all sides by
mountains, on which are situated the towns of Caxa-
marca (at 9362 feet above the sea), Bogota (8728 feet),
and 'Mexico (7469 feet); and in the Himalaya the
valley of Cashmere, 5820 feet above the sea.

Less nearly related to the " craters of elevation " than
the above described simplest form of volcanic activity,
are the numerous caldron-like depressions, surrounded
by but little raised margins which they have them-
selves formed, which are found among the extinct
volcanoes of the Eifel, in non-volcanic rock, Devonian
schist, and which are called "Maars." "They are as
it were mine-funnels, — evidences of mine-like explo-
sions,"— recalling the singular phenomenon described
by myself at the earthquake of Riobamba (4th of
February 1777), when the bones of men were cast forth
on the hill of La Culca. (314) Where, in particular
cases, such ancient explosion-craters situated at no great
elevation, in the Eifel, in Auvergne, or in Java, are
filled with water, they may bear the name of Lake-
craters, which has been given to them ; but that term
ought not to be regarded as synonymous with the
German "Maar," inasmuch as on the summits of the
highest volcanoes, on true cones of elevation, in extinct
craters, (for example, on the Mexican volcano of Toluca
at 12,245 feet, and on Mount Elbourz in the Caucasus
at 19,716 feet,) small lakes were found by myself and

230 REACTION OF THE INTERIOR OF THE EARTH

by Abich. In the Eifel volcanoes, two kinds of volcanic
activity, of very different age, must be carefully distin-
guished from each other, the strictly so-called volcanoes,
sending forth streams of lava, and the feebler explosion
phenomena of the Maars. To the first belong the
basaltic, oliviniferous, vertically columnar lava-stream
in the Uesbach valley near Bertrich ; (315) the volcano of
Grerolstein, situated in a limestone containing dolomite,
embedded in the Devonian grauwacke schist ; and
the long ridge of the Mosenberg (1753 feet above the
sea) not far from Bettenfeld, west of Manderscheid.
The last-named volcano has three craters, of which the
first and second (the northernmost ones) are perfectly
round, and their floors covered with turf-bog ; but from
the third (southernmost) (316) crater there has descended
a considerable reddish brown stream of lava, which
lower down, towards the valley of the Little Kyll, has
become columnar. A remarkable phenomenon, generally
speaking quite foreign to lava-yielding volcanoes, is, that
neither at the Mosenberg, nor at the Grerolstein, nor in
the other volcanoes proper of the Eifel, are the lava
streams seen to be surrounded at their origin by a
trachytic rock, but on the contrary, so far as they are
accessible to observation, they seem to come directly
from the Devonian beds. The surface of the Mosen-
berg in no way betrays what is hidden in its depths.
The augitic scoriae, which pass connectedly into basaltic
streams, contain small burnt pieces of schist, but no
trace of enclosed trachyte. Nor are any such enclosures
to be found even in the crater of the Kodderberg, though
it is so near to the greatest mass of trachyte in the
neighbourhood of the Ehine, the Siebengebirge.

ON ITS EXTEUIOR. YOLC ANDES. 231

" The Maars," as is acutely remarked by Captain von
Dechen, " seem by their formation as if they belonged to
nearly the same epoch as streams of lava from true
volcanoes. Both are situated near deeply cut valleys.
The lava-yielding volcanoes were certainly active at a
period when the valleys had already received very
nearly their present form ; and we actually see the
oldest lava-streams of the district falling precipitously
into the valleys." The Maars are surrounded by frag-
ments of Devonian schist and of heaped-up grey sand
and tufa margins. The Laacher See, — whether we
regard it as a great Maar, or as it is regarded by my
friend C. von Oeynhausen as part of a great caldron-
valley in the clay-slate, — shows on its surrounding
borders some volcanic scoriae ; as at the Krufter Ofen, at
the Veitskopf and the Laacher Kopf. But it is not only
by the entire absence of streams of lava (as they are
observed in the Canary Islands, at the outer margin
of true craters of elevation, or in their immediate
neighbourhood), or by the inconsiderable height of
their margin, that the Maars are distinguished from
craters of elevation ; their borders are also wanting
in that regular, always convex, outward slope in the
stratification, which is a consequence of upheaval. As
has been already remarked, the Maar depressions in
the Devonian schists appear like mine-funnels into
which, immediately after the violent explosion of hot
gases and vapours, the greater part of the light loose
masses emitted (Rapilli) again fell back. I name here, as
examples, only the Immerath Maar and those of Pulver
and Meerfeld. In the midst of the first, of which the
dry floor at two hundred feet deep is cultivated, are

232 REACTION OF THE INTERIOR OF THE EARTH

situated the two villages of Upper and Lower Immerath.
Here are found in the volcanic tufa around, just as at
the Laacher See, mixtures of felspar and augite form-
ing spheres or nodules, in which black and green
vitreous particles are interspersed. Similar balls of
mica, hornblende and augite, full of vitrifactions are
also found in the tufa margin of the Pulver Maar,
(near Gillenfeld), which, however, is entirely transformed
into a deep lake. The regularly circular Meerfelder
Maar, covered partly with water and partly with bog, is
geologically distinguished by the vicinity of the three
craters of the great Mosenberg, from the southernmost
of which a lava-stream has flowed. The Maar is, how-
ever, 640 feet lower than the long ridge of the volcano,
and is at its northern end ; moreover it is not in the
axis of the line of craters, but is more to the north-west.
The mean height of the Maars of the Eifel above the
sea falls between 922 feet, the height of the Laacher
See (if it be regarded as a Maar), and 1588 feet, the
height of the Mosbrucher Maar.

As this is more particularly the place at which to call
attention to the general uniformity and accordance in
the substances produced by volcanic activity, while
there is the greatest diversity in the forms of the out-
ward framework belonging to that activity, (as the Maars
above spoken of, — elevation craters, — and conical
volcanoes open at the summit,) I will here refer to the
striking abundance of crystallised minerals which the
Maars brought to the surface at their first explosion, and
which now lie partly buried in the tufa. This abundance
is particularly great round the Laacher See ; but there
QI e also others, for instance, the Immerather and the

OX ITS EXTERIOR. VOLCANOES. 233

Meerfelder Maar, (the latter of which is rich in balls of
olivine,) which contain many fine crystalline masses. We
will cite here Zircon, Hauyne, Leucite, (317) Apatite,
Nosean, Olivine, Augite, Ryacolite, common Felspar (Or-
thoclase), glassy Felspar, Mica, Sodalite, Grarnet, and
Titanite. If the number of fine crystallised minerals at
Vesuvius is much greater, (Scacchi counts 43 kinds,)
it should not be forgotten that very few of them were
emitted from the volcano; that the greater part of
them belong, according to Leopold von Buch's opinion,
(318) " not to Vesuvius, but to a tufa-covering, extending
far beyond Capua, which was upheaved by the cone
of Vesuvius together with itself at its elevation, and
was probably the product of a submarine volcanic action
hidden deep in the interior of the earth."

Certain determinate directions of the different kinds
of phenomena produced by volcanic activity may also be
distinctly traced in the Eifel. " The eruptions of lava
in the high Eifel took place along a fissure nearly
28 miles long, running from south-east to north-
west, from Bertrich to the Goldberg near Ormond ;
on the other hand, the Maars from the Meerfelder
Maar to Mosbruch and the Laacher See follow a direc-
tion from south-west to north-east. These two leading
lines of direction intersect each other in the three
Maars of Daun. Eound the Laacher See no trachyte is
visible at the surface. The presence of this rock in the
depths below is only indicated by the peculiar nature of
the whole felspathic pumice of Laach, as well as by the
ejected balls of augite and felspar. The only visible
trachytes of the Eifel, however, composed of felspar with
large crystals of hornblende, are distributed between

234 REACTION OF THE INTERIOR OF THE EARTH

basaltic mountains or hills; as in the Seilberg (1892 feet
high), near Quiddelbach, on the Height of Struth near
Kelberg, and in the wall-like ridge of Eeimerath near
Boos."

Next after the Lipari and Ponza Islands few parts of
Europe have produced more pumice than this part of
Germany, which, with a comparatively small elevation,
presents such different forms of volcanic activity, Maars
or craters of explosion, basaltic hills, and lava-emitting
volcanoes. The principal mass of the pumice is between
Niedermendig and Sorge, Andernach and Eiibenach ;
the principal mass of the " Duckstein " or Trass (a very
recent conglomerate deposited by water) is in the
Brohlthal, from its mouth at the Rhine up to Burg-
brohl, near Plaidt and Kruft. The trass formation of
the Brohlthal contains, besides fragments of graywacke
schist and pieces of wood, fragments of pumice which
are in no way distinguishable from the superficial
covering of the district, and even of the trass itself. I
have always doubted, notwithstanding some analogies
which the Cordilleras might seem to present, whether
the trass could be ascribed to eruptions of mud from
the lava-emitting volcanoes of the Eifel. I rather con-
jecture, with H. von Dechen, that the pumice was
thrown out dry, and that the trass was formed in the
same manner as other conglomerates. "There is no
pumice in the Siebengebirge, and the great eruption of
pumice in the Eifel (the principal mass of which still
overlies the Loess, and in parts alternates with it) may,
according to the conjecture to which the local relations
tend to lead us, have taken place in the valley of the
Rhine above Neuwied, in the great Neuwied basin,

ON ITS EXTERIOR. VOLCANOES. 235

perhaps near Urmits, on the left side of the Ehine.
•From the great friability of this substance, the later
action of the river may have entirely removed all traces
of its place of origin or eruption. In the whole line of
the Eifel Maars, as well as in the Eifel volcanoes, from
Bertrich to Ormond, no pumice is found. That of the
Laacher See is confined to the hills of its shores, and at
the other Maars the small pieces of felspar, lying in the
volcanic sand and tufa, do not pass into pumice."

We have already alluded to the relations which the
age of the Maars, and that of the very different eruptions
of lava, bear to the age of the formation of the valleys.
" The trachyte of the Siebengebirge seems much older
than the formation of the valleys, and even than the
Ehenish browncoal. Its origin is unconnected with
the rending open of the Ehine valley, even supposing
we should ascribe the formation of that valley to the
opening of a fissure. The valley is much more recent
in its origin than the Rhenish browncoal, and more
recent than most of the Ehenish basalt; on the other
hand, it is older than the volcanic eruptions of lava-
streams, and older than the great outburst of pumice,
and than the trass. Basaltic formations decidedly come
down to a more recent period than do those of trachyte ;
and the principal mass of the basalt is thus to be looked
upon as younger than the trachyte. On the present
sides of the valley of the Ehine many basaltic groups
(those of Unkeler Steinbruch, Eolandseck, and Godes-
berg) were, it may be supposed, first laid bare by the
opening of the valley, as it is probable that they were
previously enclosed in the Devonian graywacke hills."

The Infusoria (whose general distribution over our

236 REACTION OF THE INTERIOR OF THE EARTH

globe, on continents, in the greatest depths of the ocean,
and in the highest strata of the atmosphere, as demon-
strated by Ehrenberg, constitutes one of the most
brilliant discoveries of our age) have in the Eifel their
principal seat in the rapilli, trass-beds and pumice-
conglomerates. Silicious-shelled organisms abound in
the Brohlthal and in the erupted masses of Hochsim-
mern ; sometimes in the trass they are mingled with un-
carbonised branches of Coniferae. The whole of this
minute organic life is, according to Ehrenberg, a fresh-
water formation ; it is only exceptionally, in the upper-
most deposit of the friable yellowish Loess, at the foot
and on the sides of the Siebengebirge, that marine Poly-
thalamia are found, indicating the brackish character of
an ancient coast district. (319)

Is the existence of the phenomenon of Maars confined
to Western Germany ? Count Montlosier, who was well
acquainted with the Eifel by his own observations in
1819, and recognised the Mosenberg as one of the finest
volcanoes which he had ever seen, regards (like Rozet)
the Grouffre de Tazenat, the Lac Pavin, and the Lac de
la Grodivel, in Auvergne, as Maars, or explosion-craters.
They are hollowed out of very different kinds of rock,
granite, basalt, and domite (trachytic rock), and are sur-
rounded at the margin by scoriae and rapilli. (3-°)

The framework or scaffolding, which the more
powerful eruptive activity of volcanoes may be said to
construct by the upheaving and uplifting of the ground,
and by the streams of lava poured forth, has been
observed under at least six different forms, and these
different forms are found distinctly repeated in the most
distant regions of the earth. To a native of a volcanic

ON ITS EXTEKTOE. YOLCAXOES. 237

district, born amidst basaltic and trachytic mountains,
these once familiar forms, when seen again in a strange
land, often seem, as it were, to greet him with a home-
like welcome. The forms of mountains are among the
most important of the elements which determine the
"physiognomy of nature;" they give to the aspect of
the region in which they are seen, according as they are
adorned with vegetation or soar aloft in desert naked-
ness, a cheerful, or a grave and majestic character. I
have very lately tried to place side by side in a separate
atlas a number of sketches, taken from drawings of my
own, of the Cordilleras of Quito and Mexico. As basalt
appears sometimes in cones rounded off at the summit,
sometimes as twin mountains placed close to each other
but of unequal height, sometimes as a long horizontal
ridge having two higher rounded summits at either
extremity ; so in the trachyte we distinguish pre-
eminently the majestic dome (321) of Chimborazo (21,422
feet), a form not to be confounded with that of the also
unopened, but more slender, " bell-shaped " mountains.
The conical form is seen in the greatest degree of per-
fection (322) in Cotopaxi (18,877 feet) ; and next to it in
Popocateptl (323) (17,726 feet), as seen from the beautiful
shores of the lake of Tezcuco, or from the top of the
ancient Mexican terraced pyramid of Cholula ; and
in the volcano (324) of Orizaba (17,374, or, according
to Ferrer, 17,879 feet). A strongly truncated cone-
shape (3'5) is seen in the Nevado de Cayambe-Urcu
(19,365 feet) immediately under the equator, and also
in the volcano of Tolima (18,128 feet), as seen rising
from behind the primeval forest, from the foot of the
Paramo of Quindiu, near the little town of Ibague. (326)

238 REACTION OF THE INTERIOR OF THE EARTH

The volcano of Pichincha (15,890 feet), to the surprise
of the geologist, forms a long-drawn ridge, at one
extremity of which, a little higher than the rest of
the mountain, is situated the wide and still burning
crater. (327)

Fallings in of the crater-walls, occasioned by great
volcanic spasms of activity, in which they are rent
asunder by mine-like explosions from the depths below
them, give rise in conical mountains to very strange and
contrasted forms : it has been thus in the case of the
Carguairazo (15,666 feet), where a fissure into "double
pyramids," of more or less regularity of form, took place
by a sudden falling in on the night of the 19th of July
1698 ; (3-8) and in that of the finer pyramids of Ilinissa,
(17,437 feet) (329). In a similar manner the upper por-
tion of crater-walls may become broken into a rude kind
of battlement, and subsequent commotions may leave
parts only of the same homogeneous mountain standing
as partially detached towers or peaks ; and thus at Capac-
Urcu, Cerro del Altar, now only 17,456 feet, two very
similar peaks, rising in emulation of each other, allow us
to conjecture the earlier primitive form of the mountain.
A tradition has been generally preserved among the
natives of the highlands of Quito, between Chambo and
Lican, and between the mountains of Condorasto and
Cuvillan, that fourteen years before the invasion of
Huayna Capac, son of the Inca Tupac Yupanqui, the
summit of the last-named volcano, after eruptions which
lasted uninterruptedly for seven or eight years, fell in,
covering the whole of the plateau on which New Rio-
bamba is situated with pumice and volcanic ashes.
Originally higher than Chimborazo, it was called in the

ON ITS EXTERIOR. VOLCANOES. 239

Inca or Quichua language, " Capac," i. e. King or
Prince of Mountains (urcu), because its summit was
seen to rise higher than any other above the lower limit
of the snowy region. (33°) Mount Ararat, whose summit.
17,080 feet high,*was reached by Friedrich Parrot in
1829, and by Abich and Chodzko in 1845 and 18oO,
forms, like Chimborazo, an unopened dome. Its large
streams of lava broke forth far below the snow line. An
important feature in the shape of Ararat is formed by
the deep lateral hollow of Jacob's Vajley, which may be
compared to the Val del Bove in Mount Etna. It is
here that, according to Abich, the internal structure of
the nucleus of the trachytic bell-shaped mountain is
first visible, as this nucleus and the elevation of the
whole mountain are much more ancient than the streams
of lava. (331) The mountains of Kasbegk and Tschegem,
which rise from the same leading Caucasian ridge, run-
ning E.S.E. to W.N.W. as does Elburuz, or Elbourz
(19,716 feet high), are also cones without craters on their
summits: on the summit of the last-named colossal
mountain there is a crater-lake.

As cones and domes are in all parts of the world
much the most frequent forms, it is the more surprising
to see, as a solitary instance in the group of volca-
noes around Quito, the long ridge of the volcano of
Pichincha. I occupied myself long and carefully with
examining its form, and besides the view of its profile
founded on many measurements of angles, I have
published also a topographic sketch of its cross-
valleys. (332) Pichincha forms a wall more than eight
geographical miles long, consisting of black trachytic
rock (composed of augite and oligoclase), upheaved over

240 REACTION OF THE INTERIOR OF THE EARTH

a fissure in the westernmost Cordillera nearest to the
Pacific Ocean, without the direction of the axis being
coincident with that of the Cordillera. The lofty
wall is surmounted, castle-like, by three successive
summits from south-west to north-east, called Cuntur-
guachana, Gruagua-Pichincha (or the child of the old
volcano) and el Picacho de los Ladrillos. The proper
volcano is termed " the Father," or " Old Man," " Eucu-
Pichincha." It is the only part which enters the
region of perpetual snow, rising to a height which ex-
ceeds that of Gruagua-Pichincha (the child), by about
190 or 200 feet. Three tower-like rocks surround
the oval crater, which lies a little to the south-west,
therefore, out of the immediate axis, of a rocky wall, of
which the mean height is 15,673 feet. In the spring of
1802 I reached the summit of the eastern tower in
company with the Indian Felipe Aldas. We stood
alone there, on the outermost margin of the crater,
about 2450 feet high above the floor of the fiery abyss.
Sebastian Wisse, to whom the physical sciences owe so
many interesting observations made during his long
residence in Quito, had the boldness, in 1845, to pass
several nights in a part of the crater of Eucu-Pichincha,
where, towards sunrise, the thermometer fell 3°*6 below
the freezing point, or to 28°'4. The crater is divided
into two parts by a ridge of rock covered with vitrified
scoriae. The eastern part is more than a thousand feet
lower than the western one and is now the especial seat
of volcanic activity. There, a cone of eruption rises to
a height of nearly 270 feet, and is surrounded by more
than seventy burning fumaroles exhaling sulphureous
vapours. (333) It was probably from this eastern circular

ON ITS EXTERIOR. YOLC ANDES. 241

crater (which now at its least-heated places is covered
with tufts of reed-like grasses, and of a Pourretia with
leaves resembling those of bromelias), that the eruptions
of burning scoriae, pumice, and ashes from the volcano
proceeded in 1539, 1560, 1566, 1577, 1580 and 1660.
The city of Quito was then often veiled for days in
thick darkness by the falling ashes.

Of this more rare, elongated form of volcanoes, we
know, in the old world, Gralungung, with a large crater,
in the western part of Java (334), the doleritic mass of
the Schiwelutsch in the Kamtschatka, a chain whose
highest points attain 10,167 feet (335), and Hecla, which
seen from the north-west, or in the direction normal to
that of the principal longitudinal fissure over which it
was upheaved, presents a broad mountain ridge having
different small " horns " or summits. Since the most
recent eruptions, in 1845 and 1846, which gave a stream
of lava eight miles long and at some places two miles
broad, comparable to the stream of lava from Etna in
1669, there are on the back of Hecla five caldron-
shaped craters arranged in a row. As the principal
fissure is directed N. 65° E., the volcano, as seen from
Selsundfiall, i.e. from the south-west side, and therefore
as a cross section, appears a pointed cone. (336)

If the shapes of volcanoes differ so strikingly from
each other, as Cotopaxi and Pichincha, without the
emitted substances and the chemical processes of the
profound interior being altered, the relative position of
the cones of elevation is sometimes yet more singular.
In Luzon, one of the Philippine Islands, the still active
volcano of Taal, whose most destructive eruption took
place in 1754, rises in the middle of a large lake in-

YOL. IV. R,

242 REACTION OF THE INTERIOR OF THE EARTH

habited by crocodiles, called the Laguna de Bombon.
The cone, which was ascended in Kotzebue's voyage
of discovery, has a crater lake, out of which again rises
a cone of eruption with a second crater. (337) This
description recalls involuntarily the mention, in the
journal of Hanno's voyage, of an island enclosing a
small lake, in the middle of which a second island rises.
Two such instances would seem to have been met with,
one occurring in the Grulf of the Western Horn, and
the other in the Bay of the Gorilla Apes on the west
coast of Africa. (338) Descriptions so particular would
tend to make one believe them based on actual observa-
tion of nature I

The least and the greatest height of points at which
the volcanic activity of the interior of the earth mani-
fests itself permanently at the surface, is a hypsometric
consideration interesting to physical geography, — to
which belong all facts relating to the reaction of the
fluid interior of the planet against its surface. The
measure of the upheaving force (339) is, indeed, mani-
fested in the height of volcanic cones ; but great caution
is necessary before pronouncing any judgment as to the
influence of relative height upon the frequency and
strength of eruptions. Particular contrasts between
very high and very low volcanoes, in frequency and
strength of effects, can determine nothing ; and of the
several hundred active volcanoes supposed to exist on
continents and islands, our knowledge is still so incom-
plete, that the only decisive method, that of mean
values, cannot yet be regarded as applicable ; and, even
if such mean numbers could give us a definite result by
assigning the approximate altitude at which eruptions

ON ITS EXTERIOR. VOLCANOES. 243

recur oftenest, there would still remain room for doubt
how far the influence of height, or distance from the
volcanic hearth, was modified by all the incalculable
accidental circumstances, which may render the network
of fissures more or less impeded, or more or less easily
traversed. The phenomenon is then, in respect to causal
connection, " indeterminate."

Keeping cautiously to the facts only, where the com-
plexity of the phenomena, and the want of historical
information respecting the number of eruptions in the
course of centuries, do not yet permit us to arrive at
laws, I will content myself, in regard to the comparative
hypsometry of volcanoes, with giving a table of five
groups, in which the different classes of elevation are
represented by a small number of well-authenticated
examples. I have included in these five groups only
detached mountains having craters still burning, there-
fore, active volcanoes strictly so called, not unopened
domes, as Chimborazo. All cones of eruption are ex-
cluded which are dependent on an adjacent volcano, or
which, at a distance therefrom, as in the Island of
Lancerote, and in the Arso at the Epomeo in Ischia,
have not preserved any permanent communication be-
tween the interior and the atmosphere. According to
the testimony of that most zealous investigator of all
that belongs to Mount Etna, Sartorius von Walter shausen,
that volcano is surrounded by nearly 700 greater and
lesser cones of eruption. As the measured heights of
all summits are taken in relation to the level of the
sea (the present liquid surface of our planet), it is
important here to remind the reader, that island
volcanoes — of which some, as the Japanese volcano

244 REACTION OF THE INTERIOR OF THE EARTH

Kosima (34°), described by Horner and Tilesius, at the en-
trance of the Tsugar Straits, do not rise 1000 feet above
the waters, while others, as the Peak of Teneriffe (341),
are 12,000 feet above the sea — have been actually
upheaved by the volcanic forces from a sea-bottom often
above 21,000 feet, and sometimes even nearly 46,000
feet deep. It is also important to remark, in order
to avoid erroneous inferences from the subjoined nu-
merical relations, that if between volcanoes of the 1st
and the 4th class, i.e. between volcanoes of 1000 and of
18,000 French feet (1066 and 19,184 English feet), the
differences appear very considerable, yet the propor-
tion between these numbers becomes greatly altered
if (in conformity with Mitscherlich's experiments on
the fusion-temperature of granite, and according to the
not altogether probable hypothesis that the heat in-
creases uniformly in arithmetical progression with in-
creasing depth) we assume the upper limit of the
molten interior of the earth to be about 121,500 feet
below the present level of the sea. The differences of
height between known volcanoes are, we may well be-
lieve, not considerable enough to authorise us to expect
that the height of the highest should present any
materially increased obstacle to the effects of the tension
of elastic vapours (powerfully increased when pent up
by the stoppage of volcanic fissures) in forcing lava and
other dense substances up to the crater.

ON ITS EXTERIOR. VOLCANOES. 245

Hypsometrical Table of Volcanoes.

First group of volcanoes, from 700 to 4000 French
feet (from 746 to 4263 English feet) * :—

The volcano of the Japanese island Kosima, south
of Jezo : 746 feet, according to Horner.

The volcano of the Lipari Island, named "Vol-
cano : " 1304 feet, according to Fr. Hoffmann. (342)

Gunung Api (signifying, in the Malay language,
fire-mountain), the volcano of the island of Banda :
1948 feet

The volcano of Izalco (343), in a state of almost
constant eruption; first ascended in 1770; in the
State of San Salvador, in Central America: 2132
feet, according to Squier.

Grunung Ringgit, the lowest volcano in Java : 2345
feet, according to Junghuhn. (344)

Stromboli: 2957 feet, according to Fr. Hoffmann.

Vesuvius, the Eocca del Palo, on the highest
northern margin of the crater : the mean of my two
barometric measurements (346), in 1805 and 1822,
gives 3997 feet.

The volcano of Jorullo, which broke forth, in a
high Mexican plain (346), on the 29th Sept. 1759:
4265 feet.

* It would have been easy, and might have appeared more simple, to
have made the divisions.of this table at 4,000, 8,000, 12.000, and 16,000
English feet ; but this would have introduced confusion in the subse-
quent reasoning, for example, by throwing Etna and the Peak of
Teneriffe into different groups.

246 REACTION OF THE INTERIOR OF THE EARTH

Second group, from 4000 to 8000 French feet (4263
to 8526 English) : —

Mont Pele de la Martinique : 4706 feet ? according
to Dupuget.

SoufYriere de la Guadeloupe : 4867 feet, according
to Charles Deville.

Grunung Lamongan, in the eastern part of Java :
5339 feet, according to Junghuhn.

Grunung Tengger ; of all the volcanoes of Java, that
which has the largest crater (347) : height of the cone
of eruption, Bromo, 7546 feet, according to Junghuhn.

Volcano of Osorno in Chili : 7549 feet, according
to Fitz Roy.

Volcano of the island Pico (348), one of the Azores :
7613 feet, according to Vidal.

The volcano of the Isle of Bourbon : 8001 feet,
according to Berth.

Third group, from 8000 to 12,000 French feet (8526
to 12,789 English) : —

The volcano of Awatscha (in the peninsula of
Kamtschatka) ; not to be confounded (349) with the
rather more northerly Strieloschnaia Sopka, which.
English sailors commonly call the volcano of Awat-
scha: 8910 feet, according to Erman.

Volcano of Antuco (35°), or Antoio (in Chili) : 8918
feet, according to Domeyko.

Volcano of Fogo, one of the Cape de Verde Is-
lands (351): 9152 feet, according to Charles Deville.

Volcano of Schiwelutsch, Kamtschatka : north-
eastern summit, 10,549 feet, according to Erman. (352

;

ON ITS EXTERIOK. 'VOLCANOES. 247

Etna (353) : according to Smyth, 10,871 feet.

Peak of Teneriffe : 12,158 feet, according to Charles
Deville.(354)

Grunung Semera ; this volcano is the highest moun-
tain in Java: 12,235 feet, according to Junghuhn's
barometric measurement.

Mount Erebus, in S. lat. 77° 32'; the nearest
known volcano to the south pole (355) : according to
Sir James Clark Eoss, 12,366 feet.

Volcano Argseus (356) inCappadocia, nowErdschisch-
Dagh, S.S.E. of Kaisarieh : 12,600 feet, according to
Pierre von Tschichatscheff.

Fourth group, from 12,000 to 16,000 French feet
(12,789 to 17,052 English) : -

Volcano of Tuqueres(357) in the highlands of the
Provincia de los Pastes : according to Boussingault,
12,821 feet.

Volcano of Pasto (358) : according to Boussingault,
13,450 feet.

Volcano Mauna Roa(359): according to Wilkes,
13,758 feet.

Volcano of Cumbal(360) in the Provincia de los
Pastos. 15,618 feet, according to Boussingault.

Volcano Kliutschewsk (3C1) (Kamtschatka) : accord-
ing to Erman, 15,763 feet.

Volcano Kucu-Pichincha : according to barometric
measurements by Humboldt, 15,923 feet.

Volcano Tungurahua : according to a trigonome-
trical measurement by Humboldt (3ti2), 16,491 feet.

248 REACTION OF THE INTERIOR OF THE EARTH

Volcano of P.urace(363) near Popayan : 17,006 feet,
according to Jose Caldas.

Fifth group. From 16000 to above 20000 French
feet (17,052 to above 21,315 English) : —

Volcano Sangay, south-east of Quito: 17,124 feet
according to Bouguer and La Condamine. (364)

Volcano Popocatepetl (365) : according to a trigono-
metrical measurement by Humboldt, 17,726 feet.

Mount Elias (36G) west coast of North America :
according to the measurements of Quadra and Gale-
ano, 17,851 feet.

Volcano of Orizaba (367) 17,879 feet, according to
Ferrer.

Volcano of Tolima (368) : according to a trigono-
metrical measurement by Humboldt, 18,129 feet.

Cotopaxi (369) : 18,877 feet, according to Bouguer.

Volcano of Arequipa (37°) : , according to a trigono-
metrical measurement by Dolley, 18,879? feet.

Volcano Sahama (in Bolivia) (371) : according to
Pentland, 22,349 feet.

The volcano with which the fifth group terminates
is more than twice as high as Etna, and five and a half
times as high as Vesuvius. The graduated scale of vol-
canoes which we have passed in review — beginning with
the low Maars of the Eifel (mine-funnels, without
external framework, which have thrown out balls of
olivine surrounded by half-fused schistose fragments)
and ending with the still active Sahama, more than
22,000 feet high — has taught us that there is no

ON ITS EXTERIOK. VOLCANOES. 249

necessary connection between the maximum of elevation
and a less degree of volcanic activity, or the nature of
the visible rock. Observations limited to particular coun-
tries might easily lead us into error in this respect.
For example, in the part of Mexico which lies within
the tropics, all the mountains which are covered with
perpetual snow (therefore, the culminating points of the
whole country) are volcanoes ; and this is also for the
most part the case in the Cordilleras of Quito, provid-
ing we permit ourselves to associate with the volcanoes
trachytic domes not open at the summit (Chimborazo
and Corazon): but on the other hand, in the eastern
chain of the Andes of Bolivia, the greatest mountain
heights are completely unvolcanic. The Nevados of
Sorata (21,288 feet) and Illimani (21,148 feet) consist
of graywacke slates, which have been broken through
by masses of porphyry (372), in which (as still remaining
evidences of the fact) fragments of slate are found
enclosed. Also in the eastern Cordillera of Quito,
south of the parallel of 1° 35', the high snow-covered
summits of Condorasto, Cuvillan, and the Collanes,
which are opposite to the trachytic mountains, consist
of mica slate and slaty quartz. According to what
we now know (thanks to the meritorious labours of
Brian Hodgson, Jacquemont, Joseph Dalton Hooker,
Thomson, and Henry Strachey) of the mineralogical
constitution of the greatest elevations in the Himalaya,
it would appear that there also, what used to be called
the primitive rocks — granite, gneiss, and mica slate —
are the visible rocks, but not any trachytic formations.
Pentland, in Bolivia, found fossil shells in the Silurian
schists on the Nevado of Antacaua, 17,478 feet above

250 REACTION OF THE INTERIOR OF THE EARTH

the sea, between La Paz and Potosi. The enormous
height to which fossils found by Abich in Daghestan,
and by myself in the Cordilleras of Peru (between
Gruambos and Montan), show the chalk formation to
have been lifted, forcibly remind us that non-volcanic
sedimentary strata, full of organic remains, not to be
confounded with volcanic beds of tufa, show themselves
in places where, for a great distance around, mela-
phyres, trachytes, dolerites, and other pyroxenic rocks, to
which the upheaving impelling forces are attributed, re-
main concealed deep below. What vast tracts of the
Cordilleras and of the country to the eastward may be
explored without any visible trace of any granitic rock !
As I have already, more than once, remarked, the
frequency of eruptions in a volcano appears to be de-
pendent on varied and very complicated causes; no
general law can be established respecting the ratio of
absolute height to frequency and intensity of igneous
action. If in one small group, the comparison of Strom-
boli, Vesuvius, and Etna, might mislead us to suppose
that the number of eruptions is inversely proportional to
the height of the volcano, we soon find other facts which
are in direct contradiction to this supposition. Sartorius
von Waltershausen, who has done so much for the
knowledge of Etna, remarks that, taking the average of
recent centuries, an eruption may be expected about
every six years, whereas in Iceland (where, properly
speaking, no part of the island can be said to be secure
from destruction by subterranean igneous action), Hecla,
which is 5755 feet lower than Etna, has, so far back as
our knowledge extends, sent forth eruptions only every
seventy or eighty years. (373) The Quito group of

ON ITS EXTERIOR. VOLCANOES. 251

volcanoes presents a far more striking case. The vol-
cano of Sangay, 17,052 feet high, is much more active
than the little volcano of Stromboli (2957 feet); of
all known volcanoes it is the one which shows the
greatest number of eruptions of far-shining fiery scoriae,
in short intervals of time. Instead of bewildering our-
selves amidst hypotheses on the causal relations of inac-
cessible phsenomena, we may here, in preference, pause
to consider six points on the earth's surface, which are
peculiarly important and instructive in the history of
volcanic activity : — Stromboli ; the Chimsera in Lycia ;
the ancient volcano of Masaya ; the very recent one of
Izalco ; Fogo, in the Cape de Verde Islands ; and the
colossal Sangay.

The Chimsera in Lycia, and Stromboli, the ancient
Strongyle, are the two igneous phsenomena of volcanic
activity, whose historically proved permanence reaches
farthest back. The conical hill of Stromboli, a dolerite
rock, is twice as high as the volcano in the island of
Volcano (Hiera, Thermessa), whose last great eruption
was in 1775. The uninterrupted activity of Stromboli
was compared by Strabo and Pliny with that of Lipari,
the ancient Meligunis ; but " its flame " (i. e. its emitted
scoriae) is said to possess " with less heat, greater purity,
and a stronger light." (374) The number and forms
of the little fiery orifices are very variable. Spallan-
zani's representation of the crater-floor, which was long
regarded as exaggerated, has been fully confirmed by an,
experienced geologist, Friedrich Hoffmann, and more
recently by an acute physicist, A. de Quatrefages. One
of the red-glowing fiery orifices has an opening of only
2 1 feet diameter : it resembles the shaft of an oven

252 REACTION OF THE INTERIOR OF THE EARTH

or furnace ; and looking down upon it at any time from
the margin of the crater, one sees the fluid lava mount-
ing and overflowing. The permanent eruptions of Strom-
boli still serve, as in ancient times, as a guide to mariners ;
and also, by the direction of the flame and of the as-
cending vapours, afford, as they did to the Greeks and
Romans, uncertain prognostications as to weather. The
various signs of a near change of wind are connected
with the myth of Eolus's first residence at Strongyle,
and still more with observations on the then vivid fire on
Volcano (the " sacred isle of Hephaestos" — Vulcan), by
Polybius, who shows a remarkably exact knowledge of
the state of the crater. The frequency of the fiery ap-
pearances has in recent times shown some irregularity.
The activity of Stromboli, as well as that of Etna, is,
according to Sartorius von Waltershausen, greatest in
November and the winter months. It is sometimes
interrupted by pauses of repose ; but these, as the ex-
perience of many centuries has shown, are of short
duration.

The Chimsera in Lycia, which has been so excel-
lently described by Admiral Beaufort, and of which
I have already twice spoken, (375) is not a volcano, but
a perpetual fire-fountain, or a spring of gas kept always
in a state of ignition by the volcanic activity of the
interior of the earth. It has been recently visited by
a talented artist, Albert Berg, for the purpose of
obtaining picturesque views of a locality famed from
high antiquity (since the times of Ctesias and Scylax
of Caryanda), and to collect specimens of the rocks
from amidst which the flame of the Chimsera issues
forth. The descriptions of Beaufort, Edward Forbes,

ON ITS EXTERIOR. VOLCANOES. 253

and Spratt, in the "Travels in Lycia," are perfectly
confirmed. An eruptive mass of serpentine rock tra-
verses the dense limestone, in a ravine which ascends
from south-east to north-west. At the northern extre-
mity of the ravine the serpentine is cut off, or perhaps
only concealed, by a curved crest of calcareous rock.
The specimens of serpentine brought home are partly
green and fresh, and partly brown and weathered. In
both, diallage is clearly recognisable.

The volcano of Masaya, (376) which was celebrated even
at the beginning of the 16th century, and concerning
which, under the name of the Infierno de Masaya, reports
were made to the Emperor Charles V., is situated between
the lakes of Nicaragua and Managua, south-west of
the charming Indian village of Nindiri. For centuries
it presented the same rare and curious phenomena as
those described at the volcano of Stromboli. In look-
ing down from the margin of the crater, the waves of fluid
lava, as moved by the subterranean vapours, were seen to
rise and fall in the red-glowing abyss. The Spanish
historian Gonzalez Fernando de Oviedo first ascended
Masaya in July 1529, and drew comparisons between it
and Vesuvius, which he had previously visited (1501) in
the suite of the Queen of Naples, as her xefe de guarda-
ropa. The name " Masaya " belongs to the Chorotega
language of Nicaragua, and signifies " burning moun-
tain." This volcano, surrounded by an extensive lava-
field (mal-pays), which it has probably formed, was then
reckoned as belonging to the mountain-group of the
" nine burning Maribios." " In its ordinary state,"
says Oviedo, " the surface of the Java, on which black
scoria3 are floating, remains several hundred feet below

254 REACTION OF THE INTERIOR OF THE EARTH

the margin of the crater ; but sometimes, by a sudden
vehement boiling up, it almost reaches the upper rim."
The perpetual light seen from a distance is remarked by
Oviedo (who expresses himself on the subject with
great distinctness and sagacity) to have been caused, not
by any real flame, (377) but by vapours illuminated from
beneath. The light is said to have been so strong,
that at a distance of more than three leagues, on the
route from the volcano to Granada, the illumination of
the country around almost equalled that of the full
moon.

Eight years after Oviedo's visit, the volcano was
ascended by the Dominican monk Fray Bias del Castillo,
who entertained the foolish notion that the fluid lava in
the crater was liquid gold, and associated himself with
an equally covetous Franciscan monk of Flanders, Fray
Juan de Grandavo, for obtaining it. These two men,
availing themselves of the credulity of the Spanish
settlers, set on foot a " company " who were to furnish
the necessary funds, " they themselves, however," as
Oviedo sarcastically remarks, " claiming exemption," as
ecclesiastics, " from any pecuniary contribution." The
account of the execution of this bold undertaking, given
by Fray Bias himself in a report to the Bishop of Cas-
tilla del Oro, Thomas de Verlenga, has been only re-
cently made known (1840) by the discovery of Oviedo's
memoir on Nicaragua. Fray Bias del Castillo is the
same person that is spoken of in the writings of Gromara,
Benzoni, and Herrara, as Fray Bias del Inesta. He had
once served on board a vessel as a sailor, and wished to
imitate the method he had seen used by the inhabitants
of the Canary Islands for gathering the Lichen Roccella

ON ITS EXTERIOR. VOLCANOES. 255

(which furnishes a dye-stuff), from the face of precipitous
cliffs, suspending themselves for that purpose by ropes
overhanging the sea. For some months various contri-
vances were tried with a crane and pulley for supporting
a long beam over the abyss. The Dominican, with his
head covered with an iron helmet, and a crucifix in his
hand, and three other members of the association were
let down, and remained for a whole night on a part of
the solid crater floor, from whence they made vain
attempts to fill earthen vessels, lowered in an iron
caldron, with the supposed liquid gold. In order not
to discourage the shareholders (378), they agreed to say,
when they were drawn up, that they had found great
riches ; and that " el Infierno de Masaya " would
henceforth deserve to be called " el Paraiso de Masaya."
The operation was subsequently repeated several times,
until the governador of the neighbouring city of Gra-
nada, either suspecting the deceit or that the royal
revenue would be defrauded, forbade " persons being let
down by ropes into the crater." This was in the summer
of 1538; but in 1551, Juan Alvarez, Dean of the
Chapter of Leon, received from Madrid the very
naive permission "to open the volcano and take out
the gold contained in it." Such were the popular
beliefs in the 1 6th century ; and even in our own times,
in 1822, it was necessary that Monticelli and Covelli, at
Naples, should demonstrate by chemical experiments
that the ashes of Vesuvius, from the eruption of
the 28th of December, did not contain any gold ! (379)

The volcano of Izalco, which is situated 'on the
west coast of Central America, thirty-two miles north
of San Salvador, and east of the harbour of Sonso-

256 REACTION OF THE INTERIOR OF THE EARTH

nate, broke forth eleven years later than the volcano
of Jorullo in the interior of Mexico. Both outbreaks
took place in cultivated plains, and after several
months of earthquakes and of subterranean " roarings "
(bramidos). In the Llano de Izalco a conical hill was
upheaved ; and contemporaneously with its upheaval a
stream of lava began to pour forth from its summit.
This was on the 23rd of February 1770. It is still
uncertain how much of its rapidly increasing height
should be attributed to " upheaval of the ground," and
how much to " accumulation " of erupted scoriae, ashes,
and masses of tufa ; but it is certain that, instead of
soon becoming extinct like Jorullo, the volcano of Izalco
has continued, ever since its first appearance, in uninter-
rupted activity, and often serves as a lighthouse to
mariners making the land in the Bay of Acajutla. Four
eruptions of flame are reckoned to take place in an
hour ; and the great regularity of the phenomenon has
astonished the few accurate observers who have wit-
nessed it. (38°) The strength of the eruptions varied,
but not the time at which each took place. The height
attained by the volcano of Izalco was estimated after
the last great eruption, of 1825, at about 1600 feet,
nearly equal to the height of Jorullo above the original
cultivated plain from which it rose, but almost four
times as great as the crater of elevation in the Phle-
grsean Fields, called the Monte Nuovo, to which Scac-
chi (381) assigns, by exact measurement, 43 1|- feet. The
permanent activity of the volcano of Izalco, which was
long regarded as a safety-valve for the country round
San Salvador, did not, however, preserve the town from
complete destruction on Easter night in 1854.

ON ITS EXTERIOR. VOLCANOES. 257

The one of the Cape de Verde Islands which rises be-
tween S. lago and Brava had early received from the
Portuguese the name of Ilha do Fogo, because, like
Stromboli, it incessantly emitted fire ; and it continued
to do so from 1680 to 1713. After a long interval of
repose, this volcano was rekindled in the summer of
1798, a short time after the last lateral outbreak of the
Peak of Teneriffe in the crater of Chahorra, which crater
was erroneously named, as if it had been a distinct
mountain, the " Volcano of Chahorra."

The most active of all the volcanoes of South America,
and indeed of all those which I have here enumerated,
is Sangay; also called Volcan de Macas, because the
remains of the ancient and, at the time of the Conquista,
populous city of that name are situated only twenty-
eight miles to the south of it, on the Eio Upano. This
colossal mountain, 17,124 feet high, has been elevated
on the eastern declivity of the Eastern Cordillera, be-
tween two systems of tributaries to the Great Kiver of
the Amazons, i.e. the river-systems of the Pastaza and
of the Upano. The great an4 incomparable igneous
phenomenon now presented by this volcano would appear
to have only begun in 1728. In the measurement of
an arc of the meridian by Bouguer and La Condamine
(1738 to 1740), Sangay served them as a perpetual fire
signal. (382) I myself heard its thunder for months in
1802 (particularly in the early morning) from Chillo
near Quito, the pleasant country seat of the Marques
de Selvalegre, as, half a century before, Don Jorge Juan
had heard " the ronquidos del Sangay " from a point
rather more to the north-east, near Pintac at the foot of
Antisana. (383) In the years 1842 and 1843, when the

VOL. iv. g

258 REACTION OF THE INTERIOR OF THE EARTH

eruptions were combined with the greatest noise, the
sound was heard distinctly, not merely in the harbour
of Gruayaquil, but also farther to the south, along the
shores of the Pacific, as far as Payta and San Buena-
ventura ; a distance about equal to that between Berlin
and Basle, the Pyrenees and Fontainebleau, or London
and Aberdeen. Although, since the beginning of the
present century, the volcanoes of Mexico, New Granada,
Quito, Bolivia, and Chili have been visited by geologists,
yet, unfortunately, Sangay (which exceeds Tungurahua
in height), owing to its solitary position remote from all
routes of communication, has remained entirely neg-
lected. It was not until December 1849, that an adven-
turous and highly informed traveller, Sebastian Wisse,
who had resided five years among the Andes, ascended
Sangay and almost reached the highest summit of the
steep snow-covered cone ; he determined chronome-
trically the exact intervals *of the astonishingly fre-
quent eruptions, and examined the composition of
the trachyte which, limited to so restricted an area,
breaks through the gneisp ; 267 eruptions were counted
in an hour, each lasting on an average 13 "'4, (384)
and — which is very remarkable — not accompanied
by any sensible shaking of the cone of cinders. The
erupted substances, veiled for the most part in much
smoke (sometimes of a grey, and sometimes approaching
to an orange colour), consisted chiefly of mingled black
ashes and rapilli, but partly also of scoriae, which
rose perpendicularly and were globular, about 16
or 17 inches in diameter. In one of the strongest
eruptions, Wisse counted from 50 to 60 glowing stones
simultaneously erupted. Most of them fall back into
the crater ; sometimes they cover its upper margin, or

ON ITS EXTERIOK. VOLCANOES. 259

slide down a part of the cone, and are seen, at night,
shining afar in the darkness : it was this probably which
gave occasion to La Condamine's erroneous belief of
"burning sulphur and asphalte teing poured forth."
The stones are thrown up individually and successively,
so that some are falling back while others are rising. By
exact measurement by time, the visible space of fall
(reckoning therefore to the margin of the crater) was
determined in the mean to be only 785 feet. At Etna,
the stones thrown up reach, according to the measure-
ments of Sartorius von Waltershausen and the astronomer
Dr. Christian Peters, a height of 2660 feet above the
crater. Gremellaro's estimations, during the Etna erup-
tion of 1832, were even three times as high ! The
black ejected ashes form, on the slopes of Sangay and
for twelve miles around, beds 300 or 400 feet thick.
The colour of the ashes and rapilli gives to the upper
portion of the volcano a peculiarly dismal and fearful
aspect. Before closing this paragraph, I would once
again call the reader's attention to the size of this great
volcano (six times the height of Stromboli), because it
militates strongly against the belief of the lower volcanoes
having always the most frequent eruptions.

Even more important than the form and height of
volcanoes is their grouping, because it leads us to the
great geological phenomenon of elevation over fissures.
Such groups of volcanoes, whether they have been
elevated, according to Leopold von Buch, in lines, or
simultaneously around a central volcano, indicate the
parts of the earth's crust in which (whether it may
have been from the lesser thickness of the rocky strata,
or from their nature, or from their original fissuring)
the tendency of the molten interior to break forth has

260 REACTION OF THE INTERIOR OF THE EARTH

met with least resistance. Three degrees of latitude
are included in the space in which the volcanic ac-
tivity manifests itself fearfully in Etna, in the ^Eolian
Isles, in Vesuvius, and the Phlegraean Fields, from
Puteoli (Dicaearchia) to Cumae and to the fire-vomit-
ing Epopeus on Ischia, the Tyrrhenian Ape's Island
^Enaria. Such a connection of analogous phenomena
could not escape the notice of the Greeks. Strabo
says, "The whole sea, beginning from Cumae to Si-
cily, is traversed by fire, and has undoubtedly in its
depths hollow passages communicating with each other
and with the mainland. (385) Such an inflammable
nature, as is described by all, shows itself not only in
Etna, but also in the country around Dicsearchus and
Neapolis, and around Baiae and Pithecusse." Hence
arose the fable that Typhon lies under Sicily, and that,
when he turns himself, flames and water burst forth,
and sometimes even small islands and boiling water.
"Often, between Strongyle and Lipara (in this wide
sweep), flames are seen to issue from the surface of the
sea, when the fire opens for itself a passage from the
cavities in the depths, and violently forces its outward
way." In Pindar (386), the body of Typhon is so vast,
that " Sicily and the sea-surrounded heights above
Cumae (Phlegra, the 'field of burning') lie on the
monster's shaggy breast."

Thus Typhon (the raging Enceladus) became in the
Greek popular phantasy, the mythic designation of the
unknown cause of volcanic phaenomena, lying deeply
buried in the bosom of the earth. By the situation
and space assigned to his bulk, they indicated the
boundaries and connected action of the particular vol-

ON ITS EXTERIOK. VOLCANOES. 261

canic system. In the richly imaginative geological
picture of the interior of the earth in Plato's grand
contemplation of Nature, in the Phasdo (pp. 112-
114), this connected action is, with still greater boldness,
extended to all volcanic systems. In it the lava-streams
draw their supplies from the Pyriphlegethon, where,
" after it has often rolled round and round beneath the
earth," it pours itself into Tartarus. Plato says expressly,
that in " the fire-vomiting mountains, where such are
found on the earth, small portions of the Pyriphlegethon
are blown out." (OVTO? S' scrlv ov sTrovopd^ovcn Hvpi-
^>\sysdovra, ov /cal ol pva/css diroa-Trdo-^ara dvatyva-OHTiv,
OTTTJ av TV-^WCTL rrjs 777$.) The expression (page 113, B.),
driving out with violence, may be understood to refer
to the motive force of the previously enclosed and sud-
denly and forcibly escaping wind, on which subsequently
Aristotle, in his "Meteorology/' founded his whole theory
of volcanic action.

In accordance with these views, thus common to
ancient and modern times, the grouping of volcanoes is
even more characteristically marked in linear arrange-
ment, than when they are found around a single central
volcano. It is most striking where it depends on the
situation and extent of very long and mostly parallel
fissures. Thus, in the New Continent, to cite only the
most important series having the most numerous mem-
bers, we find the linear groups of Central America and
Mexico, of New Granada and Quito, of Peru, Bolivia,
and Chili ; and in the Old Continent the groups of the
Sunda Isles (the southern Indian Archipelago, espe-
cially Java), the peninsula of Kamtschatka and its
continuation in the Kurile Islands, and the Aleutian

262 REACTION OF THE INTERIOR OF THE EARTH

Islands which form the southern boundary of the almost
closed Behring's Sea. We will pause to consider some
of these leading groups in more detail ; particularities
often lead, by their intercomparison, to the fundamental
relations of phenomena.

The line of volcanoes of Central America (according
to the older nomenclature, the volcanoes of Costa Eica,
Nicaragua, San Salvador, and Guatemala) extends (from
the volcano Turrialva near Cartago to that of Soconusco)
over six degrees of latitude, 10° 9' to 16° 2' N., in a
generally S. E. and N. W. line, and, with its few curves,
has a length of 540 geographical miles, about equal to
the distance between Vesuvius and Prague. The most
near to each other (as if they had been upheaved upon
a single fissure only sixty-four miles long), are the eight
volcanoes which lie between the Laguna de Managua
and the Bay of Fonseca, between the volcano of Momo-
tombo, and that of Conseguina whose subterranean
thunder in 1835 was heard in Jamaica and in the high-
lands of Bogota, resembling the fire of artillery. In
Central America, and in the whole southern portion of the
New Continent, or, to speak even more generally, from
the Archipelago de los Chonos, in Chili, to the most
northern volcanoes of Edgecombe, on the small island
near Sitka, (387) and Mount Elias in Prince William's
Sound — over a distance of 6400 geographical miles — the
volcanic fissures have everywhere been opened on the
western side, or nearest to the shore of the Pacific. At
the point where the Central American line of volcanoes
(in lat. 131°, north of the Grulf of Fonseca) enters the
State of San Salvador, at the volcano of Conchagua, the
directions, both of this line and of the western coast-line,

ON ITS EXTERIOK. VOLCANOES. 263

change simultaneously. The line of volcanoes now runs
E.S.E. — W.N.W. ; and farther on, where they are so
near to each other that five, still in a state of greater or
less activity, are found in the short course of 120 miles, the
direction even becomes almost E. — W. This deflection
corresponds to a great widening of the continent towards
the east, in the peninsula of Honduras, where the coast-
line turns as suddenly, and runs from Cape Gracias a
Dios to the Grulf of Amatique for 300 miles due east
and west, its direction having been before, for a similar
distance, north and south. At the group of the high
volcanoes of Guatemala, 14° 10' N., the line resumes its
earlier N. 45° W. bearing, and maintains the same to
the Mexican frontier towards Chiapa and the isthmus of
Huasacualco. North-west from the volcano of Soco-
nusco to that of Turtla, there has not been found even an
extinct cone of trachyte ; mica slate and granite rich in
quartz everywhere prevail.

The volcanoes of Central America do not crown the
adjacent mountain-chains, but rise for the most part
singly, and entirely detached from each other, along the
foot of the chains. The greatest elevations are at the
two extremities. At the southern end, in Costa Eica,
from the summit Irasu (the volcano of Cartago) both
oceans are visible, to which indeed, besides the height
(11,079 feet), the more central position contributes.
South-east of Cartago there are mountains between
11,000 and 12,000 feet high,— Chiriqui 11,262, and the
Pico Blanco 11,737 feet. Nothing is known of the
rocks of which they consist; probably they are un-
opened trachytic cones. Farther to the south-east, the
mountains in Veragua sink to little more than 5000

264 REACTION OF THE INTERIOR OF THE EARTH

atid 6000 feet. This appears to be the mean height of
the volcanoes of Nicaragua and San Salvador; but towards
the north-western extremity of the whole line, not far
from the new city of Guatemala, two volcanoes again
rise to above 13,000 feet. The maxima therefore fall,
according to my attempt at hypsometric classification
of volcanoes, into the third group, like Etna and the
Peak of Teneriffe; whereas the greater number of
elevations, situated between the two extremities of the
line, are scarcely 2000 feet higher than Vesuvius. The
volcanoes of Mexico, New Granada, and Quito belong
to the fifth group, and mostly exceed 16,000 French
feet (17,052 English).

Although in Central America the continent, from the
Isthmus of Panama, through Veragua, Costa Rica, and
Nicaragua, to the parallel of 11^-° N., widens considera-
bly, yet, from the great area of the lake of Nicaragua,
and the low level of its surface (less than 130 feet above
either sea), (388) there is such a depression of the land
that an overflowing current of air from the Caribbean
Sea often passes across to the Pacific, causing those
north-east storms, dangerous to navigation, which some-
times rage uninterruptedly for four or five days. These
storms have the remarkable feature of being usually
accompanied by an almost cloudless sky. They are
termed " Papagayos," and have given this name to the
part of the west coast of Nicaragua which is between
Brito, or Cabo Desolado, and Punta S. Elena (from 11°
22' to 10° 50' N. lat.) : it is called G-olfo del Papagayo,
and, south of the Puerto de San Juan del Sur, encloses
the little bays of Salinas and S. Elena. In sailing from
Guayaquil to Acapulco, I had an opportunity of observ-

ON ITS EXTEEIOK. VOLCANOES. 265

ing the Papagayos for more than two entire days (9 —
11 March 1803) in all their strength and peculiarity,
but a little more to the south, viz. in less than 9° 13' N.
The waves were higher than any that I had ever seen ;
and the constant visibility of the sun in the clearest
blue sky made it possible to measure their height by
altitudes of the sun taken on the crest of the wave, and
in the trough of the sea, by a method which had not
then been previously tried. All Spanish, English (389),
and American navigators ascribe these storms in the
Pacific to the north-east trade in the Atlantic.

In a new work (39°) prepared with much pains, partly
from already published, and partly from manuscript
materials, on the " Reihen-Vulkane " of Central Ame-
rica, I have enumerated twenty-nine volcanoes whose
former, or present, greater or less degree of activity
may be safely affirmed. The natives give a number
more than one third greater, adducing, in so doing,
several ancient basins of eruption which may have been
only lateral eruptions on the declivity of one and the
same volcano. Among the isolated conical and bell-
shaped mountains which are there called volcanoes,
there may indeed be many which consist of trachyte
or dolerite, but which have always remained "un-
opened" and have never shown any igneous activity
since their upheaval. Eighteen may now be regarded
as " burning ; " of those which hav sent forth flames,
scoriae, and lava in the present century (1825, 1835,
1848, and 1850), seven; and in the latter part of the
last century (1775 and 1799), two. (391) The absence
of streams of lava in the great volcanoes of Quito has in
modern times repeatedly given occasion to statements

266 REACTION OF THE INTERIOR OF THE EARTH

implying a similar absence of lava in those of Central
America. It is true, indeed, that in the majority of
cases, as, for example, at the present time in the volcano
of Izalco, eruptions of scoriae and ashes are unac-
companied by streams of lava; but the descriptions
given by eyewitnesses of lava-pouring eruptions in four
volcanoes — Mndiri, el Nuevo, Conseguina, and San
Miguel de Bosotlan — militate against the supposition
above noticed. (392)

I have purposely dwelt at some length on the pecu-
liarities of the situation and near vicinity to each other
of the volcanoes of the Central American line, in the
hope that some geologist, who has made himself
thoroughly acquainted with the active European vol-
canoes, and the extinct ones of Auvergne or of the
Vivarais or of the Eifel, and (which is of the greatest
importance) who is competent to describe the minera-
logical composition of the rocks according to the re-
quirements of the present state of knowledge, might
feel himself incited to visit so near and accessible a
region. Much still remains to be done there, supposing
the traveller to devote himself exclusively to geognostical
researches and particularly to the mineralogical deter-
mination of the trachytes, dolerites, and melaphyres, —
to the discrimination of the originally upheaved mass
from the part which has been covered over by subse-
quent eruptions, — and to the discovery and recognition
of actual narrow uninterrupted currents of lava, which
are too often confounded with accumulations of erupted
scoriae. Conical or dome or bell-shaped mountains which
have never been opened are to be most -carefully distin-
guished from volcanoes which either now emit, or have at

ON ITS EXTERIOR. VOLCANOES. 267

any former period emitted scoriae and lava, like Vesuvius
and Etna, or scoriae and ashes only, like Pichinca and
Cotopaxi. I do not know anything which holds out a
better promise of a brilliant advance in our knowledge
of volcanic activity, in which sufficiently varied observa-
tions extending over large and connected continental
spaces are still greatly wanting. And further, supposing
collections of rock-specimens to be obtained and brought
home from several isolated true volcanoes, and from un-
opened trachytic cones, and also from the non-volcanic
masses which have been broken through by them, the
chemical analysis to which the specimens would be
submitted, and the chemico-geological inferences which
would be drawn from the analysis, would open a no less
wide and fruitful field. Central America and Java have
both the undeniable advantage over Mexico, Quito, and
Chili, of offering the most varied and the most crowded
examples of " scaffoldings " in which volcanic activity
is displayed over large areas.

At the point, in lat. 16° 2' K., where at the boundary of
Chiapa the very characteristic series of the Central Ame-
rican volcanoes terminates in the volcano of Soconusco,
a wholly different system of volcanoes, the Mexican one,
begins. The isthmus of Huasacualco and Tehuantepec,
so important for future commerce with the coasts of
the Pacific, as well as the more north-westerly State of
Oaxaca, is entirely without volcanoes, and perhaps also
without unopened trachytic cones. It is not until after
a distance of 160 miles from Soconusco, that the small
volcano of Turtla, in lat. 18° 28', rises on the coast of
Alvarado. Situated on the eastern declivity of the
Sierra de San Martin, it had a great eruption of flames

268 REACTION OF THE INTERIOR OF THE EARTH

and ashes on March 2, 1793. In consequence of having
made exact determinations of the latitudes and longi-
tudes of the giant Snowy Mountains and volcanoes in
the interior of Mexico, and while entering the culmi-
nating points on my large map of New Spain after my
return to Europe, I was led to the exceedingly re-
markable result, that there is there a parallel of vol-
canoes and of culminating points running from sea to
sea, and deviating only a few minutes on either side of
the parallel of 1 9°. The only volcanoes, and at the same
time the only summits covered with perpetual snow, —
therefore the only mountains exceeding 12,000 feet in
the country, — the volcanoes of Orizaba, Popocatepetl,
Toluca, and Colima, lie between the parallels of latitude
of 18° 59' and 19° 20', and mark at the same time "the
direction of a fissure of volcanic activity "360 miles in
length. (393) In the same east and west line (in lat. 19°
9'), intermediate between the volcanoes of Toluca and
Colima, distant 116 miles from one, and 128 miles from
the other, in a wide plain 2583 feet high, the new
volcano of Jorullo (4265 feet) was upheaved on the
14th of September 1759. The locality of this event,
viewed in relation to the situation of the other Mexican
volcanoes, — and the circumstance that the east and west
fissure I have just spoken of intersects at almost right
angles the south south-east and north north-west direc-
tion of the great chain of mountains, — are geological
phenomena no less important than are the distance from
the sea of the point at which Jorullo has been elevated,
and the evidences of its upheaval which I have repre-
sented in detail graphically, viz. the countless vapour-
exhaling hornitos which surround the volcano, and the

ON ITS EXTERIOR. VOLCANOES. 269

pieces of granite which, in a district otherwise entirely
devoid of granite for a great distance, I found imbedded
in the lava which had poured forth from the principal
volcano of Jorullo.

The following table contains the latitudes and alti-
tudes of the series of Mexican volcanoes which have been
elevated over a fissure running from sea to sea and
crossing the fissure of elevation of the great mountain-
chain.

Series from E. to \V. Geogr. Latitude,

Volcano of Orizaba ... 19° 2' 17" 17,879 feet.

Nevado Iztaccihuatl . . . 19 10 3 15,705

Volcano Popocatepetl . . 18 59 47 17,726

Volcano of Toluca . . . 19 11 33 15,168

Volcano of Jorullo ... 19 9 0 4,265

Volcano of Colima ... 19 20 0 13,003

The westward prolongation of this parallel of volcanic
activity in the tropical zone in Mexico conducts, at a
distance of 440 miles from the coast, to a group of
islands in the Pacific called Eevillagigedo, in the vicinity
of which Collnet has seen pumice floating, and perhaps,
yet farther, to the great volcano of Mouna Eoa (19° 28'
lat.) 3360 miles distant, without the upheaval of any
intervening islands.

The linear group of Quito and New Granada com-
prises a volcanic zone which extends from 2° S. to
nearly 5° N. latitude. The extremities of the area, in
which the reaction of the earth's interior against the
surface now manifests itself, are the uninterruptedly
active volcano Sangay, and the Paramo and Volcan de

270 REACTION OF THE INTERIOR OF THE EARTH

Ruiz, which became again active in 1829, and from
which Carl Degenhardt saw smoke issue in 1831 and
1833, he being, on the first occasion, at the Mina de
Santana in the province of Mariquita, and on the
second at Mormato. Proceeding southwards from Ruiz,
the next most remarkable traces of great phenomena
of eruption are shown by the truncated cone of the
volcano of Tolima (18,120 feet), celebrated on account
of its devastating eruption of the 12th of March 1595;
the volcanoes of Purace (17,006 feet) and Sotara, near
Popayan; of Pasto (13,450 feet), near the town of the
same name; of the Monte de Azufre (12,821 feet), near
Tuquerres; of Cumbal (15,618 feet); and of Chiles in
the Provincia de los Pastos. Then follow the historically
more celebrated volcanoes of what is more strictly called
the Highland of Quito, south of the equator, of which
four, viz. Pichincha, Cotopaxi, Tungurahua, and Sangay,
may certainly be regarded as still active. If, as will
be presently more fully shown, to the north of the
mountain-knot of Robles, near Popayan, where the
great chain of the Andes forks into three, it is only in
the middle cordillera, and not in the western one which
is nearest to the sea, that volcanic activity shows itself,
on the other hand, to the south of the said knot, where
the Andes form only two parallel chains, being those
so often mentioned in the writings of Bouguer and La
Condamine, the distribution is so equal that the four
volcanoes of Pastos, as well as Cotocachi, Pichincha,
Iliniza, Carguairare, and Yana-Urcu at the foot of
Chimborazo, have broken forth on the western cordil-
lera; and Imbabura, Cayambe, Antisana, Cotopaxi,
Tungurahua (over against Chimborazo, yet near the

ON ITS EXTERIOR. VOLCANOES. 271

middle of the narrow high plain), the Altar de los
Collanes (Capac-Urcu), and Sangay on the eastern one.
If we embrace the whole northernmost group of South
American volcanoes in one view, the opinion, often ex-
pressed in Quito and in some measure supported by
historic accounts, of a migration of the greater intensity
of volcanic activity to the southward, will seem to
acquire some degree of probability. It is true, on the
other hand, that in the south, by the side of the well
nigh incessantly active colossal Sangay, we find the
ruins of the " prince of mountains " Capac-Urcu, said
to have surpassed Chimborazo in height, but which in
the latter part of the fifteenth century (fourteen years
before the conquest of Quito by the son of the Inca
Tupac Yupanqui) fell in, became extinct, and has not
since^been rekindled.

The area not covered by groups of volcanoes in the
Andes, is far greater than is commonly supposed. In
the northern part of South America, from the Volcan
de Euiz and the cone of Tolima, the two northernmost
volcanoes of the New Granada and Quito series, to
beyond the Isthmus of Panama, about Costa Eica, where
the Central American series begins, there extends a
region of more than 600 geographical miles in length,
which is, indeed, often powerfully agitated by earth-
quakes, and which contains flame-emitting salses, but
in which true volcanic eruptions are not known. On
the south, a still longer tract in which there are no
volcanoes extends nearly 1000 miles, from Sangay,
the southern extremity of the New Granada and Quito
group, to Chacani near Arequipa, the commencement of
the Peru and Bolivia series of volcanoes. So compli-

272 REACTION OF THE INTERIOR OF THE EARTH

cated and so diverse must have been, within the same
chain of mountains, the circumstances and relations
on which the formation of permanently open fissures,
and unobstructed communication between the molten
interior and the atmosphere, depend. Between the
groups of trachytic and doleritic rocks through which
the volcanic forces become active, there intervene some-
what shorter tracts, in which the prevalent rocks are
granite, syenite, mica schists, clay slates, quartzose
porphyry, siliceous conglomerates, and calcareous rocks,
of which last a considerable portion (according to Von
Buch's skilful examination of the organic remains
brought home by Degenhardt and myself) belong to chalk
formations. As I have elsewhere remarked, the gradually
increasing frequency of rocks containing labradorite,
pyroxene, and oligoclase announces to the attentive
traveller that he is passing from what had so far been a
distinct zone, non-volcanic and often rich in silver found
in porphyry without quartz and full of vitreous feld-
spar, into volcanic regions still in free communication
with the interior of the earth.

The more exact knowledge which has been recently
acquired of the position and boundaries of the five
groups of volcanoes (i. e. the groups of Anahuac or
tropical Mexico ; Central America ; New Granada and
Quito ; Peru and Bolivia ; and of Chili) teaches us that
in the portion of the Cordilleras which extends from
19£° north to 46° south latitude, making, with the
curves occasioned by variations in the direction of the
axis, a length of about 5000 geographical miles, very
little more than the half (394) has volcanoes. If we
examine the distribution of the non-volcanic spaces, we

ON ITS EXTERIOK. VOLCANOES.

273

find that the maximum interval is between the third and
fourth of the aforesaid five groups, i. e. between the
groups of Quito and Peru (fully 960 miles), while the
least interval is between the first and second group, those
of Mexico and Central America. The four intervals
between the five groups are respectively 300, 630, 960,
and 540 geographical miles. The great distance be-
tween the southernmost volcano of Quito and the
northernmost of Peru occasions, at the first moment,
the more surprise from the old custom of calling the
measurement made on the highlands of Quito "the
Peruvian Arc." It is only the smaller and southern
portion of the chain of the Andes of Peru that is vol-
canic. According to lists which I have prepared after
a careful discussion of the most recent data, the follow-
ing is a general tabular view of the number of volcanoes
in each group.

Number of

Names of the five linear Groups of American Vol-
canoes between 19° 25' N. and 46° 8' S. Latitude.

Number of
Volcanoes in
each Group.

Volcanoes
considered
to be still

active.

Mexican Group (39S)

6

4

Central American Group (3D6) .

29

18

New Granada and Quito Groups(397)

18

10

Peruvian and Bolivian Group (39S)

14

3

Chilian Group (399) ....

24

13

According to this estimate, the total number of vol-
canoes in the five American groups is ninety-one, of which
fifty-six belong to the South American continent. I
reckon as volcanoes, in addition to those which are still
burning and active, mountains from which eruptions
are known to have taken place within historic times,
and also mountains which by their structure and by the

VOL. IV. T

274 REACTION OF THE INTERIOR OF THE EARTH

presence of craters of elevation and eruption, of lavas,
scoriae, pumice, and obsidians, are characterised as vol-
canoes once active, but which had become extinct prior
to traditional or historic evidence. Unopened trachytic
cones and domes, or unopened long trachytic ridges,
as Chimborazo and Iztaccihuatl are excluded. Yon
Buch, Charles Darwin, and Friedrich Naumann employ
the word volcano with a similar limitation. I term
" still active volcanoes" those which, when viewed in close
proximity, still show signs of activity in a greater or less
degree, and of which some have also had in recent
times great and well-attested eruptions. I have pur-
posely said " when viewed in close proximity," as this
last is a very important condition ; for many volcanoes
have had their still-subsisting activity denied because,
when viewed from the plain, the thin vapours which
at a great height rise out of the crater are not seen.
Thus, at the time of my American journey, it was even
denied that Pichincha and the great volcano of Mexico,
Popocatepetl, were still burning ! although an enter-
prising traveller, Sebastian Wisse (40°), counted in the
crater of Pichincha, around the great cone of eruption,
seventy still burning openings or " fumaroles, " and,
being engaged in measuring a base-line at the foot of
Popocatepetl, in the Malpais del Llano de Tetimpa, I
myself witnessed a very distinct eruption of ashes from
that volcano. (401)

In the New Granada and Quito line, about twice the
length of the chain of the Pyrenees, and in which ten
out of 18 volcanoes are still burning, we may point out
four smaller groups or subdivisions. Proceeding from
north to south, these are, the Paramo de Kuiz and the

ON ITS EXTERIOR. VOLCANOES. 275

adjacent volcano of Tolima (4° 55' N. lat., according to
Acosta) ; Purace and Sotara near Popayan (N. 2£°) ; the
Volcanes de Pasto, Tuquerres, and Cumbal (K 2° 20' to
0° 50') ; and the series of volcanoes from Pichincha near
Quito to the uninterruptedly active Sangay (from the
equator to 2° S.). This last-named subdivision of the
entire group is not particularly distinguished among
the volcanoes of the New World, either by its great
length or by the number it comprises ; nor does it, as
we now know, contain the highest summit : for Aconcagua
in Chili (S. lat. 32° 39'), 23,003 feet .high according to
Kellett, 23,909 feet according to Fitz Koy and Pentland,
as well as the Nevados of Sahama (22,349 feet), Parina-
cota (22,029 feet), Grualateiri (21,959 feet), and Poma-
rape (21,699 feet), all between 18° V and 18° 25' S.
latitude, are all believed to be higher than Chimborazo
(21,422 feet). Yet it is the volcanoes of Quito which,
among all the volcanoes of the new continent, enjoy the
most wide-spread renown; for it is to this part of
the chain of the Andes, and to these highlands of Quito,
that the remembrance of arduous labours for important
objects in astronomical, geodesical, optical, and barome-
trical operations, attaches itself, in connection with the
brilliant names of Bouguer and La Condamine. Where
intellectual associations prevail, where a multitude
of ideas have been awakened which have led simulta-
neously to the enlargement of several sciences, fame
remains long attached to the locality. Thus it has been
in the Swiss Alps with Mont Blanc, not on account of
its height, in which respect it surpasses Monte Eosa
only by 557 feet, nor on account of dangers overcome in
its ascent, but by reason of the value, number.

T 2

278 HE ACTION OF THE INTERIOR OF THE EARTH

variety of the physical and geological views which shed
a lustre around the name of de Saussure, and around the
scenes of his untiring labours. External nature appears
to us clothed with the most grandeur where her direct
visible image is blended with its reflection from the in-
tellectual mirror of the human mind.

The Peruvian and Bolivian series, also belonging
wholly to the equinoctial zone, and, according to Pent-
land, not covered with perpetual snow until the height
of nearly 17,000 feet is reached (Darwin, Journal,
1845, p. 244), attains its maximum elevation (22,350
feet) in about the middle of its length, in the Sahama
group, between 18° 7' and 18° 25' S. lat. At Arica, there
takes place a remarkable rounded inflection of the coast-
line, which corresponds to a sudden alteration in the
direction of the axes both of the chain of the Andes and
of the western volcanic line. From thence southwards
the direction of the coast, and at the same time of the
volcanic fissure, is no longer from N.W. to S.E., but from
north to south, which direction it maintains until near
the western entrance of the Straits of Magellan, over a
distance of more than two thousand geographical miles.
A map of the " Verzweigungen und Bergknoten der
Andeskette," which I published in 1831, shows many
similar correspondences between the outline of the
continent and the course of the Cordilleras near or dis-
tant. Thus, between the capes Aguja and San Lorenzo
(5J° to 1° S. lat.), both the shore of the Pacific and the
Cordilleras run from south to north, after having
so long run from south-east to north-west between the
parallels of Arica and Caxamarca ; and thus, from the
mountain-node of Imbaburu near Quito to that of De

ON ITS EXTERIOR. VOLCANOES. 277

los Eobles (402) near Popayan, the coast-line and the
Cordilleras agree in a course which is even from S. W. to
N.E. It seems difficult to determine anything as to the
geological causal connection in this frequently manifested
accordance between the outlines of continents and the
direction of adjacent mountain-chains (South America,
Alleghanys, Norway, Apennines).

Although at present in the volcanic series of Bolivia
and Chili, it is the ivestern branch of the Andes, or that
nearest to the Pacific, which shows most traces of still-sub-
sisting volcanic activity, yet a very experienced observer,
Pentland, has also discovered at the foot of the eastern
chain, more than 180 geographical miles from the sea,
a perfectly preserved but extinct crater with unmistak-
able streams of lava. This crater is at the summit of a
conical mountain near San Pedro de Cacha, in the valley
of Yucay 12,000 feet high (lat. 14° 8', long. 71° 20');
south-east of Cuzco, where the eastern snowy range of
Apolobamba, Caxabaya, and Vilcanoto stretches from
S.E. to N.W This noteworthy point (403) is marked by
the ruins of a celebrated temple of the Inca Viracocha.
The distance of this former lava-yielding volcano from
the sea is much greater than that of Sangay (which
also belongs to an eastern Cordillera), and greater than
that of Orizaba, or of Jorullo.

An interval of 540 miles without volcanoes divides the
volcanic series of Peru and Bolivia from that of Chili ;
this is the distance of the eruption in the desert of
Atacama from the volcano of Coquimbo. As already
remarked, 2° 34' farther to the south the Chilian group
of volcanoes reaches its maximum of elevation in
Aconcagua, 23,003 feet high, which is also, according to

278 REACTION OF THE INTERIOR OP THE EARTH

our present information, the maximum elevation of
all the mountains of the new continent. The mean
height of the Sahama group is 22,008 feet, or 586 higher
than Chimborazo. Then follow in rapidly diminishing
elevation Cotopaxi, Arequipa (?), and Tolima, between
18,877 and 18,129 feet. ' I give unaltered, in what look
like very exact numbers, the results of measurements
which, unfortunately, are necessarily compounded of
trigonometric and barometric determinations ; but I do
so with the view of inciting others to repetition and
correction. Of the Chilian group of volcanoes a few
only, out of the number of 24 mentioned in the table,
have had their elevations determined; and these are
mostly only the southernmost, or those between the
parallels of 37° 20' and 43° 40', from Antuco to Yantales,
and which are also the lowest, having only the inconsi-
derable elevation of from 6000 to a little more than
8000 feet. In Tierra del Fuego the summit of Sar-
miento, covered with perpetual snow, rises only (ac-
cording to Fitz Roy) to the height of 6821 feet. From
the volcano of Coquimbo to the volcano of San Clemente,
the distance is 968 miles.

For the continual activity of several of the Chilian vol-
canoes we have the important testimony of Darwin (404),
who regards Osorno, Corcovado, and Acongagua as
decidedly still active, and the evidence of Meyen,
Poppig, and Gray, who ascended Maipu, Antuco, and
Peteroa ; of Domeyko, the astronomer Grilliss, and Major
Philippi. The number of still-burning craters may be
taken at thirteen — only five less than in the Central
American group.

We now turn from these American continental

ON ITS EXTERIOR. VOLCANOES. 279

volcanic groups to the old continent, where, on the
contrary, we find the greater number of crowded groups
of volcanoes not on the mainland, but on islands. Thus,
most of the European volcanoes are in the Mediterranean,
and indeed (including the great and repeatedly active
crater between Thera, Therasia, and Aspronise) in the
Tyrrhenian and ^Egean portions of that sea ; and in Asia
the mightiest volcanoes are to the south and east of the
continent, in the greater and lesser Sunda Islands, the
Moluccas and the Philippines, in the islands of
Japan, and the Kurile and Aleutian archipelagos.

In no other part of the earth's surface are there such
frequent and such fresh traces of active communication
between the interior and the exterior of our planet, as
in that comprised between the parallels of 10° S. and
14° N. latitude, and the meridians of the south
point of Malacca and of the west point of the Papua
peninsula of New Guinea. The area of this volcanic
archipelago, washed by the Sunda, Banda, Solo,
and Mindoro seas, scarcely equals that of Switzerland.
The single island of Java contains at present a greater
number of still-burning volcanoes than does the whole
southern half of America, although it is only 544 geogra-
phical miles long, or one seventh of the length of South
America. A new and long-looked^for light on the geo-
logical constitution of Java (after the earlier and very
incomplete though meritorious labours of Horsfield,
Baffles, and Eeinwardt) has been recently afforded by
a courageous, accomplished, and unweariedly active ex-
plorer of nature, Franz Junghuhn. After a residence
of more than twelve years in Java he has comprised the
whole natural history of that island in an instructive

280 REACTION OP THE INTERIOR OF THE EAETH

work entitled, " Java, seine Grestalt und Pflanzendecke
und innere Bauart." Above 400 heights were measured
barometrically with great care ; the volcanic, conical, and
bell-shaped mountains, 45 in number, were all drawn in
profile, and, with the exception of three (405), were
all ascended by Junghuhn. Above half (at least 28)
were recognised as still burning and active; their
remarkable and very various forms have been most
clearly described, and the history of their eruptions, as
far as was possible, investigated. No less important
than the volcanic phsencmena of Java, are its sedimen-
tary rocks of tertiary formation, which previously to the
above-mentioned work were wholly unknown to us,
although they cover three fifths of the entire area of the
island, especially its southern portion. In many parts of
Java there are found, as remains of former extensive
forests, pieces, from three to seven feet long, of silicified
trunks of trees exclusively dicotyledonous. For a land in
which palms and tree ferns now grow in abundance,
this is the more remarkable, because in the miocene
tertiary rocks (the brown coal) of Europe, where arbo-
rescent monocotyledons do not now grow, fossil palms
are often found. (406) By diligently forming a collection
of impressions of leaves and of fossilised woods, Jung-
huhn has provided materials which, being ably examined
and worked up by Groppert into an " ancient flora " of
Java, have given us the first example of a fossil flora
from a purely tropical region.

The volcanoes of Java are far inferior in height to those
which we have been speaking of in America, the Chilian,
Bolivian, and Peruvian groups, and even to the less lofty
groups of Quito and New Granada, and of tropical
Mexico (23,000, 21,000, and 18,000 feet), — the highest

ON ITS EXTERIOK. VOLCANOES. 281

of the South American volcanoes being almost the height
of Etna above the highest volcanoes in Sumatra and Java.
In the latter island, the still-burning Gunung Semeru
is the culminating point of the whole Javanese series.
Junghuhn ascended it in September 1 844 ; the mean of
his barometric measurements gave 12,235 feet above the
sea, being 1 748 feet higher than Etna. At night the ther-
mometer sank to 43° Fahr. The more ancient Sanscrit
name of Grunung Semeru was Maha Meru (" the great
Meru"), recalling the Indian Mount Meru, which,
according to the Mahabharata was the mythic seat
of Brahma, Vishnu, and the seven Devarshi. (407) It
is remarkable that, as the natives of the high plain of
Quito had guessed antecedently to any measurement
that the Chimborazo surpassed in height all the other
snowy mountains in their country, so also the Javanese
knew the sacred mount Maha Meru, but a little distant
from Grunung Ardjuno (11,030 feet high), to be the
loftiest mountain in Java, although, from the less
height of their mountains, they had not the aid to
the judgment afforded by the inferior limit of perpetual
snow, or of a temporary snow-shower. (408) Next to
Grunung Semeru, which, as we have seen, is a little above
12,000 English feet high, come four other volcanoes of
from 10,727 to 11,116 English feet. These are (in
descending series) Grunung Slamat (409) or Mountain of
Tegal (11,116 feet), Gr. Ardjuno (11,030), Gr. Sumbing
(11,028), and Gr. Lavu (10,727). The heights of seven
other Javanese volcanoes fall between 9000 and 10,000
French feet (9591 and 10,657 English). These results
are the more remarkable, as it was previously supposed
that no mountain on the island exceeded 6000 French
feet.(410) Of the five groups of North and South Ame-

282 KEACTION OF THE INTERIOR OF THE EARTH

rican volcanoes, that of Guatemala (Central America) is
the only one which in 'mean height is exceeded by the
Java group. Although the Volcan de Fuego near Old
Guatemala is, according to Poggendorff's calculation,
13,109 English feet, or 874 feet higher than Gunung
Semeru, yet the rest of the Central American volcanic
series only range between 5000 and 7000 French feet,
instead of, as in Java, between 7000 and 10,000 feet.
The most lofty Asiatic volcano is to be found, however,
not in the islands, but on the continent ; for in the
peninsula of Kamtschatka the volcano of Kliutschewsk
rises to the height of 15,763 feet, or almost to that of
Kucu-Pichincha in the Cordilleras of Quito. The well-
filled volcanic range of Java, comprising more than
45 volcanoes, has its principal axis(411) in a W.N.W.
and E.S.E. direction (more exactly, W. 12° K), being
parallel, not to the longitudinal axis of Java itself, but
to the volcanic series in the eastern part of Sumatra.

The generality of direction in volcanic chains by no
means excludes the phenomenon to which attention has
been recently called in the great chain of the Himalayan
mountains, i. e. that three or four single lofty summits
may be so arranged as to constitute a short partial chain,
whose axis may form an oblique angle with the principal
axis of the entire chain. This fissure-phenomenon,
which has been observed, and in part put forward,
by Hodgson, Joseph Hooker, and Strachey, is of great
interest. (412) The short axes of secondary fissures
encounter the greater one, sometimes at almost a right
angle; and even in volcanic chains, the very highest
summits are often a little removed from the principal
axis. In the Javanese as in most volcanic series, no de-

ON ITS EXTERIOR. VOLCANOES. 283

terminate relation is observed between the height of the
mountain and the magnitude of the crater at the summit.
The two largest craters belong to Grunung Tengger and
Grunung Kaon. The first of these is a mountain of the
third class, only 8700 feet high. Its circular crater is
about 3J miles in diameter. The level floor of the
crater is 1865 feet below the highest part of the
surrounding escarpment, and consists of a smooth bed
of sand or pulverised rapilli, out of which fragments
of scoriaceous lava project here and there. Even the
enormous crater of Kirauea in Hawaii, which is filled
with glowing lava, is inferior in size (according to the
exact trigonometrical survey of Captain Wilkes and
the excellent observations of Dana) to the crater of
Grunung Tengger. In the middle of the latter there rise
four small cones of eruption, or, more strictly speaking,
four ridge-encircled orifices, terminating in deep funnel-
shaped hollows, of which one only, Bromo (the mythic
name Brahma, a word which in the Kawi has the sig-
nification of "fire," which does not appear in the
Sanscrit), is now not burning. Bromo presents the re-
markable phenomenon of having had a lake formed in
its funnel from 1838 to 1842, which Junghuhn has
shown owed its origin to the access of atmospheric waters,
warmed and acidified by the infiltration of sulphuric
vapours. (413) Next to Gunung Tengger, Grunung Eaon
has the largest crater, though its diameter is only half
as great. The view into it is awfully grand ; it is at
least 2400 feet deep ; and yet this remarkable volcano
(10,180 feet high, and which has been ascended and
most carefully described by Junghuhn) (4l4) is not even
named in Raifles's meritorious chart.

284 REACTION OF THE INTERIOR OF THE EARTH

The volcanoes of Java, like most of the linear groups
of volcanoes, present the important phenomenon, that
simultaneity of eruption is much more rarely observed in
cones adjacent to each other than in those which are far
apart. When, in the night of the llth of August
1772, the volcano Grunung Pepandajan (7034 feet high)
broke out into the most destructive eruption with which
the island of Java has ever been visited within historic
times, in the same night (llth to 12th of August) two
other volcanoes, Or. Tjerimai and Or. Slamat, which are
situated respectively 184 and 352 geographical miles in a
direct line from Or. Pepandajan(415), also became ignited.
Admitting the whole series of volcanoes to be over one
and the same hearth, yet assuredly the network through
which they communicate is so complicated, that the ob-
struction of ancient channels of vapour, or, in the course
of centuries, the temporary opening of new ones,
render conceivable a simultaneous outbreak at very
distant points. I may here recall the circumstance of
the sudden disappearance of the column of smoke from
the volcano of Pasto, on the morning of the 4th of
February 1797, when the dreadful earthquake of Kio-
bamba shook the high plain of Quito between Tungura-
hua and. Cotopaxi. (416)

The volcanoes of Java are described as having gene-
rally a ribbed character, to which I have never seen
anything similar in the Canaries, in Mexico, or in the
Cordilleras of Quito. The traveller already alluded to,
to whom we owe such excellent observations on the
structure of volcanoes, the" geography of plants, and the
psychrometric relations of humidity in Java, has de-
scribed this phenomenon with such remarkable clear-

ON ITS EXTERIOR. VOLCANOES. 285

ness, that, with the view of inciting others to new
investigations, I think it desirable to call attention to
this feature. Junghuhn says, " Although the face of a
conical volcano 11,000 feet high, Grunung Sumbing,
seen from some distance appears like an uninterruptedly
smooth incline, yet on a nearer view it is# found to
consist of separate long narrow ridges or ribs, which
become more and more divided, and wider, in descending.
They proceed from the summit, and still more
frequently from a height of a few hundred feet below
the summit, to the foot of the mountain, diverging on
all sides like the ribs of an umbrella." These long
ridges or ribs have sometimes for a short distance a
winding course; they are all divided from each other
by intervening clefts 300 or 400 feet deep, which also
widen out in descending. These radiated furrowings
of the surface are found "on the sides of all the
volcanoes in Java, but with considerable differences in
regard to their mean depth and to the distance of their
upper commencement from the margins of the craters
or the unopened summits of the different mountains.
Grunung Sumbing, 11,030 feet high, is one of those
which show the finest and most regularly formed ribs,
and with the greatest distinctness, as the mountain is
without forest trees and is covered with grass." Accord-
ing to the measurements given by Junghuhn (417), the
subdivision of the ribs increases with the decreasing
angle of inclination. Thus, in GK Sumbing, above the
zone of 9000 French feet there are only about 10 such
ribs; at 8500 Fr. feet there are 32 ; at 5500 Fr. feet,
72 ; and at 3000 Fr. feet (3197 English), above 95 ;-
the angle of inclination decreasing at the same time

286 EEACTION OF THE INTERIOR OF THE EARTH

successively from 37° to 25° and 101°. The ribs of the
volcano Gr. Tengger (8700 feet) are almost as regular ;
those of Gr. Ringgit have been in a great degree covered
over and destroyed by the devastating eruptions which
have taken place since 1586.(418) The origin of these
very peculiar longitudinal ribs and of the intervening
clefts, of which drawings are given, are ascribed to
" erosion by torrents."

The mean quantity of meteoric water in this tropical
region is indeed much greater than in the temperate
zone: the rain often falls in floods resembling water-
spouts; and although on the whole the quantity of
moisture diminishes in ascending higher in the atmo-
sphere, yet on the other hand great conical mountains
are powerful attracters of clouds, and, as I have shown
elsewhere, volcanic eruptions have a specific action in
exciting tempests. Phenomena in some degree analo-
gous are, indeed,presented by the clefts and valleys, called
barrancos, in the volcanoes of the Canaries and of the
Cordilleras of South America, described by Von Buch(419)
and myself, which are important to the geological
traveller in giving him access to the interior of the
mountain, and sometimes even conducting him near to
the highest summit and to the escarpment surrounding
a crater of elevation; but although the barrancos at
times form channels through which the meteoric waters
are carried off, yet their first origin is not to be ascribed
to such waters. (42°) It is probable that fissures result-
ing from the folding of the soft upheaved and only
later hardened trachytic masses preceded any effects
of erosion by the action of the water. But in the deep
barrancos on the declivities of domes or conical moun-

ON ITS EXTERIOR. VOLCANOES. 287

tains in the volcanic district visited by me, "en las
faldas de los Cerros barrancosos," there were no traces
of the regularity and radiating arrangement which
Junghuhn has made known to us in the singularly
formed volcanoes of Java.(421) Some analogy thereto
may perhaps be traced in the phenomenon to which
Von Buch and the ingenious observer of volcanoes
Poulet Scrope had already called attention,, viz., the
circumstance that great fissures almost always open in
the normal direction of the declivity, radiating (but
without ramification) from the centre of the mountain,
not transversely to it, either at a right or at an oblique
angle.

The belief in the entire absence of streams of lava
in the Javanese volcanoes (422), to which Leopold von
Buch was inclined as the result of the experience of the
meritorious Keinwardt, has been rendered more than
doubtful by recent observations. Junghuhn does, indeed,
remark that "the great volcano Gunung Merapi, within
the historic period of its eruptions, has not poured
forth any compact, connected streams of lava, but has
only ejected lava fragments or unconnected pieces of
rock, although, for nine months of the year 1837, lines
of fire were nightly seen descending the slope of the
cone of eruption." (423 But the same traveller has
lescribed circumstantially and distinctly three black ba-
Itic lava-streams on three volcanoes — GL Tengger,
Gr. Idjen, and Slamat.(424) On the last-named volcano
the lava-stream, after giving occasion to a waterfall, is
prolonged to the tertiary rocks. (425) Junghuhn, in
describing the eruption of Gr. Lamongan(426), July 1838,
distinguishes with great accuracy between such true out-

288 REACTION OF THE INTERIOR OF THE EARTH

pourings of lava, forming connected masses, and What he
terms a stream of stones, consisting of glowing fragments,
for the most part angular, emitted in quick succession :
" There was heard the crashing sound of the hurled-up
stones falling on the declivity and rolling down, looking
like points of fire, following each other either in line
or without order." I desire to fix attention on the very
different modes in which the appearance of fire on the
slope of a volcano may be produced, because in the dis-
cussion on the maximum angle of fall of streams of lava,
streams of glowing stones or fragments of rock (masses
of scoriae) have sometimes been confounded with conti-
nuous lava-currents.

As much has recently been said respecting the
rarity, or supposed entire absence, of lava-streams in
Java — a problem important in relation to the internal
constitution of volcanoes, and which I venture to add
has not been treated with sufficient earnestness, — the
present appears to me a suitable opportunity for con-
sidering the question in a more general point of view.
Although it is very probable that in a group or a series
of volcanoes all the members will stand in certain
common relations to the common hearth (the molten
interior), yet each individual presents peculiar physical
and chemical processes in respect to strength and
frequency of action, to degree and form of fluidity,
and to diversity of products — peculiarities not to be
explained by differences of form or of elevation above
the present surface of the sea. The colossal Sangay is as
uninterrupted in eruption as the low Stromboli. Of two
volcanoes near to each other, one sends forth pumice
without obsidian, and the other emits both together;

s

'

ON ITS EXTEBIOR. VOLCANOES. 289

one erupts only loose scorise ; the other, lava flowing in
narrow streams. Moreover in many cases these character-
istic processes do not appear to have been always the same
at different epochs. Rarity or absence of lava-streams
does not characterise either continent so as to dis-
tinguish it from the other. Particular lava-streams
may escape recognition from a variety of circumstances.
Amongst these may be noticed an overlaying covering
of tufa, ashes (rapilli), or pumice; the confluence,
either at the same or different epochs, of several
currents forming an extended lava-field, or space
covered with fragments: and in an extensive plain,
little cones of eruption, from which as at Lancerote
streams of lava may have flowed, may have been
destroyed. In the earliest conditions of our unequally
cooling planet, in the first foldings of its surface, it
appears to me very probable that there may have been
an abundant flowing out of semifluid trachytic and
doleritic masses, of pumice, and of perlite containing
obsidian, from a complicated network of fissures above
which no volcanic framework had yet arisen. The
problem of such outpourings from simple fissures de-
serves the attention of geologists.

In the Mexican series of volcanoes the greatest and,
since my American journey, the most celebrated phe-
nomenon of upheaval is the elevation of the newly
appeared Jorullo, and the emission of lava from it.
This volcano, whose topography, established by measure-
ments, I was the first to make known(427), presents by its
position, intermediate between the two volcanoes of
Toluca and Colima, and by its having broken forth over
the great fissure of volcanic activity (42S) which extends

VOL. iv. u

290 REACTION OP THE INTERIOR OF THE EARTH

from the Atlantic to the Pacific, an important and, for
that reason, a much contested geological phenomenon.
By tracing upwards the great stream of lava which had
flowed from the new volcano, I succeeded in pene-
trating deep into the interior of the crater, and iu
setting up my instruments there. , The theatre in which
this great natural event took place was a wide and
previously long-tranquil plain in the former province of
Michuacan, more than 120 geographical miles from any
other volcano ; the time was the night of the 28th to the
29th of September 1759 ; and for two full months previ-
ously, uninterrupted subterranean noises had been heard.
Unlike the wonderful "bramidos" of Gruanaxuato in 1784,
which I have described elsewhere (429), these noises were
accompanied, as is more usually the case, by earthquake
shocks, which phenomenon was altogether absent in the
neighbourhood of Gruanaxuato in 1784. The upheaval
of the new volcano, at 3 A.M. in the morning of the
29th September, was announced on the preceding day
by a phenomenon which in other eruptions usually
marks, not their beginning, but their end. On the
spot where the volcano now stands there was formerly
a thick growth of Gruava trees (Psidium pyriferum\
of the fruit of which the natives are particularly fond.
Labourers from the sugar-plantations of the Hacienda
de San Pedro Jorullo, belonging to the wealthy pro-
prietor Don Andres Pimentel, who was then living in
Mexico, had gone out to gather guavas. When they
returned to the hacienda, it was remarked, with astonish-
ment, that their wide straw hats were covered with
volcanic ashes. From this it would appear that, in
what is now called the Malpais, probably at the foot of

ON ITS EXTEEIOR. VOLCANOES. 291

the high basaltic cupola el Cuiche, fissures had opened^
and that these ashes had been emitted from them before
any change had been seen to take place in the plain.
From a letter found in the Episcopal archives of Val-
ladolid, written by Father Joaquin de Ansogorri, three
weeks after the day of the first outburst, it would
clearly seem that Father Isidro Molina (sent from the
Jesuits' College of the neighbouring station of Patz-
cuaro "to administer spiritual consolation to the in-
habitants of the Playas de Jorullo, greatly disquieted
by the subterranean noises and the shocks of earth-
quake") was the first to recognise the increasing danger,
and thus became the means of rescuing the whole of
the scanty population.

In the first hours of the night the black ashes were
already a foot high from the ground ; every one fled to
the heights of Aguasarco, an Indian village 2400 feet
above the old plain of Jorullo. From these heights
they saw (so says the traditional account) dreadful
outbursts of flame over a wide extent of country ; and
" in the midst of the flames " (according to the expres-
sion of the actual spectators of the upheaval of the
mountain) " there appeared a large shapeless lump
(bulto grande) like a black castle (castillo negro)."
In the then very sparsely inhabited state of the
country (indigo and cotton were then very little cul-
tivated) not a single human being lost his life in the
violent and long-continued earthquake — which, how-
ever, as I found by manuscript notices (43°), overthrew
houses at the copper mines of Inguaran, in the little
town of Patzcuaro, at Santiago de Ario, and for many
miles further, not, however, beyond S. Pedro Churumuco.

u 2

292 REACTION OF THE INTERIOR OF THE EARTH

At the Hacienda de Jorullo, in the confusion of the
general flight by night, a deaf and dumb negro slave
was forgotten and left behind. A Mestizo had the
humanity to return and rescue him while the house
was still standing ; and the narrators still recounted
with pleasure that he was found kneeling, with a con-
secrated taper in his hand, before the picture of Nuestra
Senora de Guadalupe.

According to the widely prevailing and accordant
native tradition, the eruption of large rocks, scoriae,
sand, and ashes was in the first days always accom-
panied by that of muddy water. In the account above
alluded to, written on the 19th of October 1759, by a
man who, with an accurate knowledge of the locality,
was describing that which had just taken place, it is
said expressly " que espele el dicho Volcan arena
ceniza y agua." All eyewitnesses relate (I translate
from the description in the official report " on the State
of the Volcano of Jorullo," March 10, 1789, by the
intendant Colonel Eiafio and the German commissary
Franz Fischer, who had entered into the Spanish ser-
vice) that "before the dreadful mountain appeared
(antes de reventar y aparecerse este terrible cerro) the
earthquake-shocks and subterranean noises increased
more and more ; that on the day itself the flat ground
was seen to rise visibly (se observe que el plan de la
tierra se levantaba perpendicularmente), and the whole
more or less swelled up so that blisters (vexigones)
appeared, of which the largest is now the volcano (de
los que el mayor es hoy el cerro del volcan). These
i-aised blisters, of very different magnitude, and some of
them of tolerably regular conical form, afterwards burst

ON ITS EXTERIOB. VOLCANOES. 293

(estas ampollas, gruesas vegigas 6 conos diferentemente
regulares en sus figuras y tamanos reventaron despues).
and emitted boiling-hot mud (tierras hervidas y ca-
lientes), as well as scoriaceous rocks (piedras cocidas ? y
fundidas), which are still to be found, covered with black
masses of stone, to an enormous distance."

These historic accounts, which could, indeed, be
wished to have been more full, agree perfectly with all
that I was able to learn fourteen years after Antonio de
Kiano's ascent. To all my questions whether the
" mountain castle " had risen gradually higher in the
course of months or years, or whether it had been seen
from the earliest days as a high summit, I could obtain
no answer. Kiano's statement, that eruptions had con-
tinued to occur for the first sixteen or seventeen years,
or up to 1776, was contradicted as untrue. The ap-
pearances of small eruptions of mud and water, in the
first days, simultaneously with the burning scoriae, were
ascribed by the tradition to the drying up of two brooks
which, rising on the western declivity of the mountain of
Santa Ines (east therefore of the Cerro de Cuiche),
watered plentifully the sugar-plantations of the former
Hacienda de San Pedro de Jorullo, and flowed on far
to the westward, to the Hacienda de la Presentacion.
There is still shown, near their supposed source, the
point where their once cold waters are said to have
disappeared in a cleft when the eastern border of the
Malpais was elevated. Eunning beneath the Hornitos,
they become warmed, and reappear, according to the
universal belief of the country-people, in two thermal
springs. As the raised part of the Malpais is there
broken off precipitously, they form two little waterfalls.

294 REACTION OF THE INTERIOR OP THE EARTH

which I have, drawn. Each has retained its former
name, Kio de San Pedro and Eio de Cuitimba. I found
the temperature of the steaming waters at this point
126°'9. The waters, during their long course, have
become warmed only, but not acidulated. The test-
papers, which I always carried about with me, were not
at all altered by them ; but farther on, near the
Hacienda de la Presentacion, towards the Sierra de las
Canoas, there gushes forth a spring impregnated with
sulphuretted hydrogen, forming a basin twenty feet
broad.

In order to obtain a clear idea of the complicated
form in relief of the space of ground in which such re-
markable upheavals took place, we must distinguish,
hypsometrically and morphologically, (1) the situation
of the volcanic system of Jorullo in relation to the mean
level of the high plain of Mexico, (2) the convexity of
the Malpais, which is covered by thousands of hornitos,
(3) the fissure, upon which six volcanic mountain-masses
have risen.

On the western declivity of the Cordillera Cen-
tral de Mexico, which runs from S.S.E. to N.N.W.,
the plain of the Playas de Jorullo (only about 2560
feet above the level of the Pacific) forms one of the
horizontal mountain-stages, or terraces, which every-
where in the Cordilleras interrupt the descent, and
therefore retard more or less the change of temperature
experienced. When from the central plateau of Mexico,
7000 French feet (7460 English) in mean height, we
descend to the wheat-fields of Valladolid de Michuacan,
to the pretty Lake of Patzcuaro with its small inhabited
island Janicho, and into the meadows around Santiago

ON ITS EXTEBIOK. VOLCANOES. 295

de Ario, which Bonpland and I found decked with the
since then so well-known flowers of the dahlia, we have
not yet lessened the altitude a thousand feet. But when
we pass from Ario, on the steep declivity above Agua-
sarco, to the level of the old plain of Jorullo, in that
very short distance we diminish the absolute height
about 4000 feet.(431) The round convex part of the
upheaved plain is about 12,800 feet across. The vol-
cano of Jorullo itself, and the five other mountains which
rose together with it upon the same fissure, are so
placed that only a small part of the Malpais falls to
the east of them. The number of hornitos is therefore
much greater on the western side ; and when in the
early morning I came out of the Indian hut in which I
had passed the night, in the Playas de Jorullo, or when
I ascended a part of the Cerro del Mirador, I saw the
black volcano rise very picturesquely above the count-
less columns of white smoke issuing from the hornitos
(the little ovens or furnaces). Both the houses of the
playas and the basaltic hill Mirador are on the level of
the ancient unvolcanic or, to speak more cautiously,
not upheaved ground. The beautiful vegetation, in
which a host of salvias bloom under the shadow of a
new kind of fan palm (Corypha pumos) and a new kind
of alder (Alnus jorullensis), contrasts with the desert
nakedness of the Malpais. The comparison of the
height of the barometer (43^) at the point where the
upheaving in the playas begins, to a point situated
immediately at the foot of the volcano, gives a differ-
ence of 473 feet. The house which we inhabited
stood about 3200 feet from the edge of the Malpais.
This edge is a small vertical precipice only twelve feet

2vV REACTION OF THE INTERIOR OF THE EARTH

high, over which the waters of the brook, which had be-
come heated (the Eio de San Pedro), fall. The best ex-
amination which I was thus enabled to make of the face of
the precipice showed black horizontal beds of mud or clay,
intermixed with sand or rapilli. At other points, which
I had not seen, Burkart observed " at the perpendicular
edge of the upheaved ground, where it is difficult of
ascent, a light-grey, much-weathered, not dense basalt
with many grains of olivine." (433) This accurate and
experienced observer (434), when on the spot, imbibed,
as I had done, the impression of a blistering of the
surface by elastic vapours, contrary to the opinion of
celebrated geologists (435), who ascribed the convexity,
found by me by direct measurement, solely to the greater
mass of the outpoured lava more immediately around
the foot of the volcano.

The many thousand small cones of eruption (or more
strictly of a roundish or oblong shape, like a baker's
oven), which are pretty equably distributed over the
upheaved surface, are generally about from four to ten
feet high. Much the greater number are on the west
side of the great volcano ; the eastern portion, towards
the Cerro de Cuiche, being scarcely one twenty-fifth of the
whole upheaved area of the playas. The hornitos are
composed of weathered balls of basalt, consisting of
concentric shells, of which I often counted as many as
twenty-four or twenty-eight. The balls are rather sphe-
roids than spheres, generally from sixteen to nineteen
inches in diameter ; but there are also many others vary-
ing between one foot and three feet. The black basalt is
pervaded by hot vapours, and has become earthy ; but
the nucleus is more dense, while the shells, when sepa-

ON ITS EXTERIOR. VOLCANOES. 297

rated, show yellow stains of oxide of iron. Singularly
enough, the soft bed of clay which unites the balls to-
gether has also a curved lamellar structure insinuating
itself into all the interstices between the balls. At
first sight I asked myself whether the whole, instead
of consisting of weathered basaltic balls containing a
little olivine, might not be masses in which a formative
process had been disturbed while taking place. This
view is opposed by the analogy of the actual hills of
basaltic balls interspersed with beds of clay and
marl which are found, often of very small dimensions,
in the Mittelgebirge of Bohemia, sometimes isolated,
and sometimes crowning both extremities of long
basaltic ridges. Some of the hornitos are either so
loosely cemented, or have such considerable internal
cavities, that, when mules are urged to pass over the
flatter ones amongst them, the animals' fore feet sink in,
whereas, in similar trials, the hills constructed by
termites (ants) do not give way.

I did not find any scoriae, or fragments of older rocks
which had been broken through, enveloped in the basalt
of the hornitos, as was the case in the lavas of the great
Jorullo. The circumstance which particularly justifies
the name of hornos, or hornitos (ovens), is that the smoke
issues from them, not at the top, but at the sides. (I speak
of the time when I saw them and wrote in my journal,
Sept, 18, 1803.) In 1780 it was still possible to light
cigars at them by fastening the cigar to a stick pushed
in two or three inches deep ; in some places the air was
so much heated by their neighbourhood, as to oblige the
passer-by to turn out of his way. Notwithstanding the
cooling-down which, according to the unanimous testi-

298 REACTION OF THE INTERIOR OF THE EARTH

mony of the Indians in the neighbourhood, had taken
place in the last twenty years, I found the temperature in
the cracks of the hornitos generally between 199°-4 and
203° ; and standing at a distance of fully twenty feet
from some of the hills, where not touched by any
vapours, the temperature of the air was still 108°*5 and
116°*2, when the proper temperature of the playas at the
same hour was barely 77°. The weak sulphuric vapours
discoloured test-paper, and for some hours after sunrise
were seen to rise to a height of from 60 to 64 feet. The
view of the columns of smoke was most striking in the
cool of the early morning. Towards noon, and even at
11 A.M., they had sunk so low as only to be visible when
quite near. In the interior of several of the hornitos,
we heard a sound like the fall or rush of water. As
already remarked, they possessed little solidity, and
when Burkart visited the Malpais, twenty-four years
after me, he found none sending forth smoke, the
temperature of most of them was no higher than that
of the surrounding air, and in many all regularity of
shape had been obliterated by the effects of rain, &c.
Burkart found near the principal volcano small cones of
a brownish-red conglomerate composed of very loosely
connected rounded or angular pieces of lava. There is
still seen, in the midst of the area covered by the
hornitos, the remains of a former rising ground, running
east and west, and drawn on my map, on which the
farm-buildings of San Pedro had rested ; its preserva-
tion at the foot of the great volcano excites astonishment.
Part of it only is covered with volcanic sand (or burnt
rapilli). The projecting basaltic rock, overgrown with
aged trunks of Ficus indica and Psidium, is assuredly, as

ON ITS EXTERIOK. VOLCANOES. 299

well as the Cerro del Mirador and the high mountain
masses whose curve bounds the plain to the east-
ward, to be regarded as having existed before the
catastrophe.

I have still to describe the great fissure over which six
volcanoes (not touching each other) have been raised, in
a direction running generally from S.S.W. to N.N.E.,
but in the three southernmost almost S. to N. The fissure
(1700 toises in length) is therefore a little curved ; its
general direction is almost perpendicular to that which
I have assigned to the volcanoes of Mexico, taken from
sea to sea. Such differences need not excite surprise,
inasmuch as the partial relations of a single group, and
those of the great masses across an entire continent,
ought not to be confounded. Thus, the long ridge of
Pichincha has not the same direction as the system of
the Quito volcanoes ; and in non-volcanic chains, e. g. in
the Himalaya, the highest points are often far from
being situated in the general elevation-line of the chain,
being instead, on shorter snowy ridges, almost at right
angles to it.

Of the six volcanic hills raised over the fissure above
spoken of, the three southernmost, between which the
route to the copper mines of Inguaran passes, are in their
present state the least interesting. They are no longer
open, and are entirely covered with greyish-white vol-
canic sand, not consisting, however, of pumice, of which,
or of obsidian, I saw no signs in this district. At Jo-
rullo, as at Vesuvius according to the statements of Von
Buch and Monticelli, the last and covering fall of ashes
seems to have been the white one. The fourth mountain
(going northward) is the great and proper volcano of

300 KEACTION OF THE INTERIOR OP THE EARTH

Jorullo, whose summit, notwithstanding its small eleva-
tion (4265 feet above the sea, 1150 feet above the
Malpais at its foot, and 1680 feet above the old surface
of the playas), was not attained without difficulty by
Bonpland, Carlos Montufar, and myself, Sept. 19th,
1803. We thought we should be most sure of reaching
the crater, then still filled with hot sulphurous vapours,
by climbing the steep side of the great lava-stream
which had burst forth from the summit itself. Our way
lay over a curled scoriaceous cauliflowerlike lava, which
gave a clear ringing sound when struck. Parts of it
have a metallic lustre ; others are like basalt, and are full
of fine grains of olivine. When we had thus gained
the top of the lava by an ascent of 721 feet in vertical
height, we turned to the very steeply inclined white cone
of cinders (on which, from its great steepness, the tra-
veller is liable to painful hurts from the sharp jagged
lava). The upper margin of the crater, on the south-
western part of which we placed our instruments, forms
a ring only a few feet in breadth. We afterwards took
the barometer down into the oval crater of the truncated
cone, where we found the temperature of the air issuing
from a cleft 200°*7. We were now 150 feet below the
margin ; and the depth of the deepest part of the hollow,
which the thick sulphurous vapours forbade our reaching,
appeared to us to be about as much again. The discovery
which interested us most was that of several sharply
defined white fragments of a rock containing felspar,
three or four inches in size, baked into the black basaltic
lava. I took them at first for syenite (43G) ; but by Cfus-
tav Eose's accurate examination of a specimen which I
brought home, they belong rather to the granite forma-

ON ITS EXTERIOE. VOLCANOES. 301

tion, which Burkart also saw crop out under the syenite
of the Eio <le las Balsas. Rose said, " the enclosure is
a mixture of quartz and felspar. The dark-green spots
appear to be not hornblende, but mica fused with some
felspar. The white baked-in fragment has been split by
volcanic heat; and in the crack, white tooth-shaped
molten threads run from one edge to the other."

More to the north than the great volcano of Jorullo,
and than the scoriaceous lava-summit which it threw
out in the direction of the old basalt of the Cerro del
Mortero, follow the two last of the six volcanoes. They
appear to have been at first very active ; for the natives
still call the extreme one the " volcancito." A wide
fissure which has opened to the westward bears in this
part traces of a destroyed crater. The great vol-
cano appears, like Epomeo in Ischia, to have poured
forth a great stream of lava only once. There is, his-
torically, no reason to suppose that it continued to do
so beyond the immediate epoch of the great outburst ;
for the letter of Father Joaquin de Ansogoni, written
twenty days after the first eruption, speaks only of
the best means to be adopted for the pastoral care of
the dispersed country-people, who had fled from the
catastrophe, and for the next thirty years we are
entirely without documentary evidence. As tradition
speaks very generally of " fires covering so great a
space," we may suppose that all the six hills over the
great fissure, and all the part of the Malpais on which
the hornitos appeared, were burning at the same time.
We may in some degree infer the heat which must have
prevailed in the surrounding atmosphere from the tem-
peratures still measured by me after the lapse of forty-

302 REACTION OF THE INTERIOR OF THE EARTH

three years : we are reminded thereby of the early state of
our planet, when under every latitude, and for long pe-
riods of time, the temperature of the atmosphere and
the distribution of organic life were subject to modifi-
cations arising from the thermic influence of the interior
acting through deep fissures.

Since my description of the hornitos surrounding
Jorullo, many analogous features in different parts of the
world have been compared to them ; they, however, still
appear to me to stand very much alone in respect to
their internal composition. If we are to give the name
of cones of eruption to all elevations emitting vapours,
the hornitos certainly deserve to be called (t fumaroles ; "
but the term " cones of eruption," applied to them,
would lead to the erroneous supposition of their showing
traces of having erupted scoriae, or even poured forth
lava, as many cones of eruption have done. Wholly
different, for example (to recall a phenomenon of greater
magnitude), in Asia Minor, on the former limits be-
tween Mysia and Phrygia, in the ancient " burnt land "
(Katakekaumene), "where it is dangerous to live on
account of earthquakes," are the three orifices which
Strabo calls " (pva-ai, " (bellows), and which have been
rediscovered by the meritorious traveller William
Hamilton. (437) Cones of eruption, such as are seen in
the island of Lancerote, at Tinguaton, or in Lower
Italy, or (only 20 feet high) on the declivity of the
Kamtschatkan volcano Awatscha (438), which was as-
cended by my friend and Siberian travelling companion,
Ernest Hoffmann, in July 1824, consist of scoriae and
ashes surrounding a little crater which has emitted
them, and which has in turn had its margin raised

ON ITS EXTERIOR. VOLCANOES. 303

by them. In the hornitos there is nothing resembling a
crater; and they consist (which is an important cha-
racter) of mere basaltic balls of concentric separable
shells, without any intermission of loose angular scoriae.
At the foot of Vesuvius, at the great eruption of 1794
(as also at earlier epochs), there were formed, arranged
over a longitudinal fissure, eight different little craters
of eruption, " bocche nuove," the so-called "parasitic"
eruption-cones, pouring forth lava, which would of it-
self distinguish them altogether from the Jorullo hor-
nitos. "Your hornitos," wrote Leopold von Buch to
me, "are not cones heaped up by the fall of erupted
substances ; they have been upheaved directly from the
interior of the earth." The same great geologist com-
pares the origin of the volcano of Jorullo itself to that
of the Monte Nuovo in the Phlegraean Fields. The
same view, in respect to the elevation of the six volcanic
mounts, impressed itself, as the most probable one (see
above, pp. 292, 293), on Colonel Riano and Commissary
Fischer in 1789, on myself, at first sight, in 1803, and
on Oberbergrath Burkart in 1827. In regard to both
the "new mountains" (Monte Nuovo and Jorullo),
which arose in 1538 and 1759, the same questions
presented themselves. Respecting the Italian Mount,
fuller accounts have been collected nearer to the time
of the catastrophe, and derived from more qualified
observers, by Falconi, Pietro Griacomo di Toledo,
Francesco del Nero, and Porzio. One of these (the
most learned), the celebrated Porzio, says, "Magnus
terrse tractus, qui inter radices montis, quern Barbarum
incolse appellant, et mare juxta Avernum jacet, sese
erigere videbatur et mentis subito nascentis figurarn

304: REACTION OF THE INTERIOR OF THE EARTH

imitari. Iste terrae cumulus aperto veluti ore magnos
ignes evomuit, pumicesque et lapides cineresque." (439)

From the geological description of the volcano of

Jorullo which I have here completed, we pass to the

more eastern part of Central Mexico. Unmistakable

lava-currents,, of which the mass is chiefly basaltic, have

been poured forth by the Peak of Orizaba, according to

the most recent and interesting researches of Pieschel

(March 1854)(440) and H. de Saussure. The rock of

the Peak of Orizaba, like that of the great volcano of

Toluca(441), which I ascended, is composed of hornblende,

oligoclase, and some obsidian, whereas the fundamental

mass of Popocatepetl is a Chimborazo rock composed

of very small crystals of oligoclase and augite. At the

foot of the eastern declivity of Popocatepetl, west of the

town la Puebla de los Angeles, I found, on the Llano de

Tetimpa (where I measured a base for determining

the heights of the two great nevados which bound the

valley of Mexico, Popocatepetl and Iztaccihuatl), an

extensive and in some respects perplexing lava-field

(or field of rough fragments of lava). It is called the

Malpais of Atlachayacatl (the latter being the name

of a low trachytic dome, on the side of which the Eio

Atlaco takes its rise); it extends from east to west,

or almost at right angles to the volcanoes, rising

abruptly from about 64 to 85 feet above the adjacent

plain. From the Indian village of San Nicholas de los

Ranchos to San Buenaventura, I estimated the length of

the Malpais at above 19,200 feet, and its breadth at

about 6400 feet. On it blocks of black lava, only

scantily overgrown here and there with lichen, of

fantastic wildness of form and position, contrast with

ON ITS EXTERIOR. VOLCANOES. 305

the yellowish-white covering of pumice which invests
all around. The pumice consists here of fragments of
coarse texture, of two or three inches diameter, in
which there are sometimes crystals of hornblende.
This coarser pumice-sand is different from the very fine-
grained one which on Popocatepetl itself, near the rock
of el Frayle and at the limit of perpetual snow, makes
the ascent so dangerous, because, when set in motion on
the steep declivity, the rolling mass of sand threatens to
overwhelm everything. Whether this malpais (the
Spanish term for these fields of lava-ruins, called in
Sicily sciarra viva, and in Iceland Odaada Hraun)
belongs to old successively superimposed lateral erup-
tions from Popocatepetl or to T the rounded conical
mountain Tetljolo (Cerro del Corazon de Piedra), I
cannot decide. It is a geologically important cir-
cumstance, that further to the east, on the route to
the little fortress of Perote (the old Aztec Pinahui-
zapan), between Ojo de Agua, Venta de Soto, and
el Portachuelo, the volcanic formation of coarse-tex-
tured white friable pearlstone (442) rises by the side of
a probably tertiary limestone (marmol de la Puebla).
This pearlstone is very similar to that of the conical
hill of Zinapecuaro (between Mexico and Valladolid),
and contains, besides laminae of mica and lumps of
imbedded obsidian, glassy bluish-grey, sometimes red-
dish, stripes resembling jasper. The extensive "pearl-
stone district" is here covered with fine-grained sand
consisting of weathered pearlstone, which at first sight
might be mistaken for granitic sand, and which, not-
withstanding their affinity in respect of origin, may
easily be distinguished from the greyish-white pumice
VOL. iv. x

306 REACTION OF THE INTERIOR OF THE EARTH

sand. The latter belongs more to the immediate vicinity
around Perote, — to the plateau, more than 7000 feet
high, which lies between the two volcanic chains, run-
ning north and south, of Popocatepetl and Orizaba.

When, in proceeding from Mexico to Vera Cruz, one
begins to descend from the heights of the non-quartz-
ose trachytic porphyry of Vigas towards Canoas and
Jalapa, one twice passes over fields of fragments of
scoriaceous lava, — the first time between the station of
Parage de Garros and Canoas or Tochtlacuaya, the second
time between Canoas and the station of Casas de la Hoya.
The first is called Loma de Tablas, on account of the
many upright basaltic slabs of oliviniferous lava, and the
second simply the Malpais. A small ridge of the same
trachytic porphyry full of glassy felspar, which (near
la Cruz Blanca and Eio Frio) forms the eastern boundary
of the Arenal (pearlstone sand-fields), divides the
Loma de Tablas from the much more extensive
Malpais. Those among the country-people who are
well acquainted with the district, affirm that the lines
of scoriae are prolonged towards the S.S.W., therefore
towards the Cofre de Perote. Having myself ascended
the Cofre and made many measurements on it(443), I
have been little disposed to infer, from what is indeed a
very probable prolongation of this stream of lava
(marked as such in my "Profiles," tab. 9 and 11, as
well in the " Nivellement barometrique "), that it had
actually flowed from that singularly shaped mountain.
The Cofre de Perote, which is indeed nearly 1400 feet
higher than the Peak of Teneriffe, but which yet is an
inconsiderable mountain compared to its colossal neigh-
bours Popocatepetl and Orizaba, forms, like Pichincha,,

ON ITS EXTERIOR. VOLCANOES. 307

a long rocky ridge, on the southern extremity of which
stands the cubical rock (la Pena) whose shape has
given occasion to the ancient Aztec name Nauhcampa-
tepetl. I found, on my ascent of the mountain, no
trace of a fallen in crater, or of lateral eruption-open-
ings; nor did I find any masses of scoriae, or any
obsidian, pearlstone, or pumice belonging to it. The
blackish-grey rock is very uniform, composed of much
hornblende and a kind of felspar which is not glassy
felspar (sanidine), but oligoclase, which would stamp
the whole of the rock which is not porous, as a dioritic
trachyte. I describe the impressions which I imbibed.
If the substances which cover this black desolate
Malpais, on which I have dwelt purposely as opposing
too partial views of volcanic manifestations of force
from the interior, have not flowed from a lateral open-
ing in the Cofre de Perote itself, yet the upheaval of
that isolated mountain, 13,550 feet high, may have
occasioned the origination of the Loma de Tablas. At
such an upheaval, longitudinal fissures and a ramifying
fissured network may have been opened for a con-
siderable distance around by the "folding" of the
surface ; and from these fissures molten substances may
have flowed directly, either as dense masses or as
scoriaceous lavas, without the intervention or formation
of the framework of open cones, or of elevation-craters.
We know that we may look in vain, in the great moun-
tains of basalt and porphyritic schist, for either central
points (crater mountains) or lower ridge-encircled
hollows, to which their common origin might be ascribed.
Science gains by the most careful distinction between
whatever is genetically different in phenomena, —

308 REACTION OF THE INTERIOR OF THE EARTH

between the formation of conical mountains with open
craters on their summits, and lateral openings, — or of
elevation-craters and maars, — and the upheaval of
closed domes or open cones, and the direct emission of
substances from fissures. The greater variety of views
necessarily called forth by a more extended horizon of
observation, and the strict critical comparison of that
which we find in nature with what was previously
supposed to be the only mode of origin, is itself a most
powerful stimulus to investigation. Even on European
ground, in the island of Eubosa rich in hot springs,
within historic times, in the great plain of Lelantum
—at a distance from any mountains, — a great stream
of lava was poured forth from a fissure. (444)

In the next group of volcanoes, the Central American
one, in which eighteen may still be regarded as active,
four (Nindiri, el Nuevo, Conseguina and San Miguel de
Bosotlan) have been recognised as yielding streams of
lava. (446) The mountains of the third group, that of
Popayan and Quito, have, for more than a century been
reputed to send forth no lava-streams, but solely masses
of unconnected glowing scoriae, ejected from the <>in>
summit-crater and often seen to glide in succession
down the mountain side. La Condamine was already of
that opinion (44G) when he left the highland of Quito and
Cuenca, in the spring of 1743. Fourteen years later.
after returning from an ascent of Vesuvius (4th June,
1755), on which he had accompanied the sister of Fre-
derick the Great, the Margravine of Bavaria, he took
occasion, at a meeting of the Academy, to express him-
self strongly as to the entire absence of proper .sin -an is
of lava (laves coulees par torrens de matieres liquefiees)

ON ITS EXTERIOR. VOLCANOES. 309

in the volcanoes of Quito. His " Journal d'un Voyage
en Italic/* read at the meeting of the 20th of April
1757, was first published in 1762, in the Memoirs of the
Paris Academy, and is of some importance in the his-
tory of the recognition of extinct volcanoes in France,
because in it La Condamine, without being aware of the
certainly earlier statements of Guettard (447), expressed
himself with great distinctness and sagacity as to the
existence of ancient crater-lakes and extinct volcanoes
in Middle and Northern Italy, and in the south of
France.

The striking contrast between the thus early recog-
nised existence of narrow undoubted streams of lava in
Auvergne, and the often too positively and unreservedly
asserted absence of any flow of lava in the Cordilleras,
engaged my earnest attention during the whole of my
expedition. My journals written at the time are con-
stantly filled with considerations on this problem,
of which I long sought the solution in the absolute
altitude of the summit, and in the height of the escarp-
ment surrounding the deep depression on cones rising
from mountain plateaus 8000 or 9000 feet above the
sea But we now know that Macas (Sangay), a scoria3-
emitting Quito volcano 17,000 feet high, is more unin-
terruptedly active than the lowly Izalco and Stromboli ;
and we also know that Antisana and Sangay on the east
have free and open declivities towards the plains of
the Napo and the Pastaza, as have Pichincha, Iliniza
and Chimborazo on the west, towards the water-sheds of
the Pacific. In many instances the upper portion rises
without any surrounding escarpment, more than 8000 or
9000 feet above the high plateau. Moreover, all these

310 REACTION OF THE INTERIOR OF THE EARTH

heights above the level of the sea (regarded, perhaps
not quite correctly, as the mean level of the earth's
surface) are inconsiderable in relation to what we may
suppose to be the depth of the seat of volcanic activity,
and of the temperature required for the fusion of rocks.
The only phenomena resembling narrow flows of lava
which I ever discovered in the Cordilleras of Quito, are
those presented by the colossal Antisana, whose height I
determined by trigonometric measurement at 19,132
feet As in this case configuration affords us the best
criterion, I will avoid the systematic expression " lava,"
as apt to convey too limited a sense in regard to origin,
and will use, in a purely objective sense terms equivalent
to the French expression "trainees de masses volca-
niques." The great mass of Antisana forms, at a height
of 13,455 feet, an almost oval plain, of which the
longest diameter is about 80,000 feet, and from which the
snow-covered portion of the volcano rises like an island.
The highest summit is rounded off into a dome, which
is connected by a short jagged ridge with a truncated
cone lying to the north of it. On this high plain,
which is partly desert and sandy, and partly covered
with grass, and on which dwell a very courageous race of
bulls (which, owing to the small amount of atmospheric
pressure, often bleed at the mouth or nostrils when
stimulated to any great muscular exertion), there is a
small hacienda, a single house, in which we passed four
days with a temperature varying between 38°*7 and
48°*2. The plain, which is not surrounded by any es-
carpment, as in craters of elevation, bears all the signs
of being an ancient lake-bottom. The Laguna Mica,
west of the Altos de la Moya, may be regarded as the

ON ITS EXTERIOR. VOLCANOES. 311

remains of a former watery covering. At the limit of
perpetual snow, the Kio Tinajillas takes its rise, which
afterwards, under the name of the Rio de. Quixos,
becomes an affluent of the Maspa, the Napo and the
Amazons. Two narrow wall-like elevations, which in
my topographical sketch of Antisana I have marked as
" coulees de lave," and which the natives call Volcau de
la Hacienda ami Yana Volcan (Yana, in the Quech-hna
language, signifies black or brown), rim like bands from
the foot of the volcano to the snow-line on the south-
western and northern declivities, and extend, as it
appears to mo, with very moderate angles of inclination
in a X.E. and S.W. direction about 13,000 feet over the
plain. With a very small breadth, they are about 192 to
213 feet above the ground of the Llanos de la Hacienda
de Santa Lucia, and del Cuvillan. Their declivities are
everywhere very abrupt and steep, even at the extremi-
ties. They consist, in their present state, of scaly and, for
the most part, sharply angular fragments of a black ba-
saltic rqck, without olivine and hornblende, but contain-
ing a few small crystals of white felspar. The mass has
often a pitchlike lustre, and contains interspersed obsi-
dian, which may be still more distinctly recognised, and
in great abundance, at Cueva de Antisana, of which we
found the height 15,942 feet. It is not, properly
speaking, a cave, but only a sort of shelter formed by
masses of rock which have fallen against each other,
which is resorted to by the herdsmen who ascend
the mountain, and in which we took refuge during a
violent hail-storm. It is a little to the north of the
Yolcan de la Hacienda. In the two narro~\\^rocky ridges,
which have the appearance of cooled lava-streams, the

312 REACTION OF THE INTERIOR OF THE EARTH

slabs and blocks are partly scoriaceous, and even puffed
up like foam at their edges, and partly weathered and
mixed with earthy detritus.

Analogous, but more complex, phenomena are pre-
sented by another, also band-like, assemblage of rocks.
On the eastern declivity of Antisana, nearly 1300 feet
lower in vertical height than the plain of the Hacienda,
in the direction of Pinantura and Pintac, there are two
small circular lakes, of which the northernmost is called
Ansango, and the southernmost, Lecheyacu. The first
has in it an island rock, and (which is very decisive) is
surrounded by rolled pumice. Each of these lakes
marks the commencement of a valley ; the two valleys
unite, and their broader prolongation bears the name of
the Volcan de Ansango ; from the margins of the two
lakes there proceed narrow lines of rocky fragments
quite similar to those above described, which do not fill
up the valleys, but rise in their middle, as dikes of 200
and 250 French feet (213 and 266 English). A glance
at the topographical plan in the " Atlas geographique et
physique" of my American journey (pi. 26) will render
these relations clearer. The blocks, here also, are partly
sharp-cornered, and partly scoriaceous at the edges, and
even have a burnt appearance. The mass is black, and
resembles basalt with scantily interspersed glassy felspar ;
some fragments are a blackish brown, and have a faint
pitchy lustre. There is, however, an entire absence
of the olivine which is so abundant on the Eio Pisque,
and at Gruallabamba, where I saw basaltic columns 72
feet high, and fully 3 feet in diameter, containing both
olivine and hornblende interspersed. In the Ansango
dike many slabs, split into that form by weathering,

ON ITS EXTEBIOR. VOLCANOES. 313

indicate porphyritic schist. All the blocks have a
yellowish grey crust from weathering. As the line of
rocky fragments (the Spanish-speaking natives call it
los derrumbamientos, la reventazon) extends from the
Eio del Molino, not far from the hacienda of Pintae, up
to the small pumice-surrounded crater lakes, the impres-
sion which naturally presented itself was, that the hol-
lows now filled with water are the orifices through which
those fragments came to the surface. A few years before
my arrival in the country, without any preceding sensible
agitation of the ground, the rocky fragments were in a
state of motion down the declivity for weeks, and some
houses at Pintac were overthrown by their impetus and
pressure. They are still without any trace of vegetation ;
which is beginning to appear, though sparingly, on the
doubtless older and more weathered blocks on the high
plain of Antisana.

What name ought we to give to the particular mani-
festation of volcanic activity, whose effects I have been
describing ? (448) Have we here to do with lava-streams ?
or only with semiscorified and glowing masses, ejected
disconnectedly but in close succession, as has been seen
to take place in very recent times on Cotopaxi ? May
it even be possible that the dykes of stones of Yana
Volcan and Ansango may have been fragmentary masses,
which, being loosely heaped up, and therefore imper-
fectly supported, in the interior of the volcano, may
have been displaced by earthquakes, and then by their
fall and shock have occasioned small local agitations,
and thus have been ejected without any renewal of
higher temperature in the volcano? Or are none of
the three very different manifestations of volcanic ac-

314 REACTION OP THE INTERIOR OF THE EARTH

tivity, which have been alluded to, applicable in this
case? and have the linear collections of rocky frag-
ments been actually elevated over fissures at the places
where we now see them (at the foot or in the vicinity of
a volcano) ? The two walls of fragments (on the high
and little inclined plain) called Volcan de la Hacienda
and Yana Volcan, and which I formerly spoke of (con-
jecturally only) as cooled lava-streams, still appear to
me, speaking from recollection at so great a distance of
time, to lend but little support to the view referred to
•in the last sentence. In the Volcan de Ansango, in
which the lines of fragments can be traced uninter-
ruptedly, like a river-bed, to the pumice-covered margins
of two small lakes, there is nothing in the angle of fall,
i.e. the difference of level between Pinantura (9477
feet) and Lecheyacu (12,150 feet), for a distance of
about 7700 toises (16,413 yards), inconsistent with the
present state of our knowledge as to the small average
angle of inclination in lava-currents. The difference of
level, 2673 feet, gives an inclination of 3° 6'. A partial
rise of the ground in the middle of the lowest part need
not be regarded as an obstacle, as cases of fluid masses
overriding such rises have been observed ; for example
in the eruption of Skapta Jokul in Iceland, in 1783
(Naumann, Greognosie, Bd. i. s. 160).

The word lava does not designate any particular
mineralogical composition of rock; and whereas Leopold
von Buch says that all is lava which flows in the inte-
rior of a volcano, and is capable by its fluidity of forming
fresh beds of deposit, I would add that fluidity is not
essential to such changes. Already, in the first de-
scription (449) of my attempt to reach the summit of

ON ITS EXTERIOK. VOLCANOES. 315

Chimborazo (first published in 1837 in Schumacher's
Astronomisches Jahrbuch), I had expressed this con-
jecture, when speaking of the remarkable pieces of
augitic porphyry which, on the 23rd June 1802, I col-
lected in loose fragments of twelve or fifteen inches
diameter at a height of 19,000 feet, on the narrow
rocky ridge leading to the summit. They were porous,
and of a reddish colour, minutely cellular, and the cells
shining. Some had a blackish tinge, and in some
cases were of very light weight, like pumice, and looked
as if freshly altered by fire. They have not, however,
flowed lava-like in currents, but have probably been
expelled from fissures on the declivity of the pre-
viously upheaved mountain dome. This explanation
of their origin would find great support in the view
entertained by Boussingault, who considered the vol-
canic cone itself to be " a heap of angular fragments
of trachyte, upheaved in a solid state, and piled upon
each other without any definite arrangement. As a
heap of broken fragments of rock occupy more space
than would the same quantity of material before being
fractured, there remain large intervening hollow spaces
in which, by mutual pressure and shock (apart from the
effect of the force of volcanic vapours), movement
arises." I am far from doubting the occurrence of
such fragments and of such hollows, which in the
Nevados became filled with water, although the fine,
regular, and generally quite vertical, columns of tra-
chyte of the Pico de los Ladrillos and Tablahuma on
the Pichincha, and, most of all, those above the little
lake of Yana Cocha, on Chimborazo, appear to me to
have been formed on the spot. My long and dearly

316 REACTION OF THE INTERIOR OF THE EARTH

valued friend Boussingault, whose chemico-geological
and meteorological views I always delight in sharing in,
regards what is called the Volcano of Ansango, and
which I now look upon rather as an eruption of frag-
ments from two small lateral craters (on the west side
of Antisana, not far from Chussulongo), as an elevation
or upheaval of blocks (45°) over long fissures. Having
examined this region, with much sagacity, thirty years
after me, he lays much stress on the analogy which the
geological relations of the eruption of Ansango to Anti-
sana appeared to him to present with those of Yana
Urcu (of which I had made a special topographical
plan) to Chimborazo. I was myself less inclined to
believe in an elevation over fissures situated imme-
diately below the whole linear extent of the Ansango
series of fragments, because, as I have said, they can
be traced up to two orifices now filled with water.
Unfragmentary wall-like elevations of great length and
uniform direction are not unknown to me, having seen
and described such in Chinese Tartary, consisting of
horizontal banks of granite. (451)

Antisana had a fiery eruption (452) in 1590, and
another in the beginning of the last century, probably
in 1728. Near the summit, on the N.N.E. side, there
is remarked a black mass of rock on which even fresh
fallen snow does not remain. For several days in the
•spring of 1801, at a time when the summit was on all
sides perfectly free from clouds, a column of black
smoke was seen to ascend from this point. On the
18th of March 1802, Bonpland, Carlos Montufar, and
I, arrived on a rocky ridge covered with pumice and
black basaltic scoriae, in the region of perpetual snow, at

ON ITS EXTERIOR. VOLCANOES. 317

a height of 18,141 feet, therefore 2358 feet higher than
Mont Blanc. The snow near the rock was, at several
places, solid enough to bear our weight, a circumstance
which is very rare within the tropics. (The temperature
of the air was from 28°'8 to 34°*5.) On the southern
declivity, by which we did not ascend, at the Piedro de
Azufre, where the rock scales off from the effects of
weathering, masses of pure sulphur, ten or twelve
feet long and two feet thick, are found : there are no
sulphur springs in the neighbourhood.

Although in the eastern Cordillera, Antisana, and
especially its western declivity (from Ansango and
Pinantura towards the little village of Pedregal), is
separated from Cotopaxi by the extinct volcano of Pas-
suchoa (453) (with its crater, la Peila, recognisable at a
great distance), by the Nevado Sinchulahua, and by the
less lofty Ruminaui ; yet there is a certain similarity in
the rocks of which these two great volcanoes are com-
posed. Beginning from el Quinche, the whole eastern
Cordillera has produced obsidian, and yet el Quinche,
Antisana, and Passuchoa belong to the basin in which
the city of Quito is situated, while Cotopaxi bounds
another basin, that of Lactacunga, Hambato, and
Riobamba. These two basins are divided from each
other by the little mountain knot of the Altos of
Chisinche, and the small dimensions of this dividing
wall renders more striking the circumstance, that the
waters of the northern side of Chisinche flow through
the Eios de San Pedro, de Pita, and de Guallabamba,
into the Pacific, while the waters of the southern side
flow through the Rio Alaques, and the Rio de San
Felipe into the Amazons and the Atlantic Ocean. The

318 REACTION OF THE INTERIOR OF THE EARTH

connection of the two Cordilleras by mountain knots
and mountain walls or dikes (sometimes low like the
Altos de Chisinche, just mentioned ; and sometimes as
high as Mont Blanc, as over the Passo del Assuay)
appears to be a more recent, and also a less important,
phenomenon, than the elevation of the parallel chains
themselves. As Cotopaxi, the greatest of all the Quito
volcanoes, presents, in its trachytic rock, many analogies
with Antisana, so also we find on the slopes of Cotopaxi,
lines of rocky fragments similar to those which have
occupied us above, and more numerous.

We felt a great desire to trace up these lines of
fragments to their origin, or rather to the place where
they are hidden under the covering of perpetual snow.
We ascended on the south-west slope of the mountain,
from Mulalo (Mulahalo), by the side of the Rio Alaques,
which is formed by the Rio de los Banos, and the Rio
Barrancas, to Pansache (12,066 feet high), where we
occupied the roomy Casa del Paramo, on the grassy
plain el Pajonal. Although much snow had fallen
sporadically at night, we succeeded, keeping to the east
of the celebrated Cabeza del Inga, in arriving, first at
the Quebrada and Reventazon de las Minas, and sub-
sequently, keeping still more to the east, over the
Alto de Suniguaicu, at the Ravine of the Lion's Mount
(Puma Urcu), where the barometric readings gave an
elevation of 14,470 feet. Another line of fragments,
which, however, we only saw from a distance, ran from
the eastern part of the snow-covered cone of cinders,
towards the Rio Negro (a tributary of the Amazons),
and towards Valle Vicioso. Whether these blocks,
which we suppose to have been glowing masses of

ON ITS EXTEBIOR. VOLCANOES. 319

scoriae, molten only at their edges, — sometimes angular
and sometimes roundish, of six or eight feet diameter,
rarely scaly, like those on Antisana, — have all been
thrown up to great heights from the summit-crater, and,
falling down on the declivity, have had their move-
ment accelerated by the rush of melted snow water ;
or whether, without having been thrown up into the
air, they have been expelled from lateral fissures of the
volcano, as the word reventazon would indicate, remains
uncertain. Returning soon from Suniguaicu and the
Quebrada del Mestizo we examined the long and broad
ridge, which, stretching from KW. to S.E., connects
Cotopaxi with the Nevado de Quelendana. Here the
blocks arranged in line are wanting, and the whole
appears a dyke-like elevation, on which the small conical
Mount el Morro rises, and nearer to Quelendana (which
has the shape of a horse-shoe,) there are several marshes
and two small lakes, the Lagunas de Yauricocha and de
Verdecocha. The rock of el Morro and of the whole
linear volcanic elevation was greenish-grey porphyritic
schist, in beds or strata eight inches thick, dipping
very regularly towards the east at an angle of 60°.
Of proper currents of lava, there was nowhere any
trace. (454)

There is an analogy between what takes place in the
Island of Lipari, rich in pumice, where on the north of
Caneto a lava-stream of pumice and obsidian descends
from the well-preserved extinct crater of the Monte di
Campo Bianco towards the sea, in which it is remarka-
ble that the fibres of the pumice are parallel to the di-
rection of the stream (455), and the local relations, which
were carefully examined by me, of the extensive pumice-

320 REACTION OF THE INTERIOR OF THE EARTH

stone quarries situated four miles from Lactacunga.
These quarries, in which the pumice-stone is in horizon-
tal layers, having quite the appearance of a rock in situ,
excited the astonishment of Bouguer in 1737. (456) He
says : " On ne trouve sur les montagnes volcaniques que
de simples fragments de pierre-ponce d'une certaine
grosseur ; mais a sept lieues au sud du Cotopaxi, dans un
point qui repond a notre dixieme triangle, la pierre-
ponce forme des rochers entiers ; ce sont des banes pa-
ralleles de 5 a 6 pieds d'epaisseur dans un espace de
plus d'une lieue carree. On n'en connoit pas la profon-
deur. Qu'on s'imagine quel feu il a fallu pour mettre
en fusion cette masse enorme, et dans 1'endroit meme
oil elle se trouve aujourd'hui; car on reconnoit aisement
qu'elle n'a pas ete derangee et qu'elle s'est refroidie dans
1'endroit ou elle a ete liquifiee. On a dans les environs
profite du voisinage de cette immense carriere ; car la
petite ville de Lactacunga, avec de tres jolis edifices, est
entierement batie de pierre-ponce depuis le tremblement
de terre qui la ren versa en 1698."

The piimice-stone quarries are situated near the
Indian village of San Felipe, in the hills of Gruapulo
and Zumbalica, which are 512 feet above the high
plain, and 9990 feet above the sea. The uppermost
layers of pumice-stone are thus 500 or 600 feet below
the level of Mulalo, the villa of the Marques de
Maenza, situated at the foot of Cotopaxi, and which was
built entirely of blocks of pumice-stone : it was once a
fine piece of architecture, but has been reduced to ruins
by frequent shocks of earthquake. The subterranean
quarries are at unequal distances from Tungurahua and
Cotopaxi, being 32 geographical miles from the former,

ON ITS EXTERIOK. VOLCANOES. 321

and only 16 from the latter. They are reached by a
gallery. The workmen assured me, that square blocks
of 20 or 21 feet could be obtained entire and without
cross cracks, from the horizontal solid beds, of which a
few are surrounded by muddy pumice detritus. The
pumice-stone, partly white and partly bluish grey, is of
very fine texture, and has long fibres of a silky lustre.
The parallel fibres have sometimes a knotted appearance,
and they then show a curious structure. The knots
are formed by roundish morsels of finely porous pumice,
about a tenth of an inch thick, round which the long
fibres curve so as to enclose them. Brownish black
mica, in six-sided small tablets, white oligoclase crystals,
and black hornblende, are scantily interspersed ; but no
glassy felspar, which elsewhere, as at Camaldoli, near
Naples, is found in pumice-stone. The pumice-stone of
Cotopaxi differs from that of the Zumbalica quarries
(457) : its fibres are short, not parallel, but confusedly
bent. Magnesia-mica is not, however, peculiar to the
pumice-stones, it is also found in the trachyte of
Cotopaxi. (458) The volcano of Tungurahua appears to
be entirely wanting in pumice. There is no trace of
obsidian in the quarries of Zumbalica, but in the blocks
thrown out by Cotopaxi, and lying near Mulalo, I have
found, in very large masses, black obsidian, having a
conchoidal fracture embedded in bluish grey weathered
pearlstone. Specimens are preserved in the Royal
Mineralogical Collection at Berlin. It would appear,
therefore, that the quarries of pumice-stone 16 miles from
the foot of Cotopaxi, which have been described above,
are not at all related to that mountain in mineralogical
constitution, and only stand towards it in the same
VOL. iv. Y

322 REACTION OP THE INTERIOR OF THE EAIITII

connection, which all the volcanoes of Pasto and Quito
bear to the volcanic hearth of the equatorial Cordilleras,
embracing an area of very many hundred square miles.
Were these pumice-stones the centre and interior of a
special crater of elevation, whose outer encircling ridge
has been destroyed in the many revolutions which the
earth's surface has undergone in the§e regions ? or were
they deposited horizontally, in apparent tranquillity, over
fissures at the time of the first crumpling or folding of
the earth's crust ? For still greater difficulties attach
to the supposition of their having been deposited as
sediment from flows of water, as is often seen in volcanic
masses of tufa, mixed with remains of plants and
shells.

Similar questions are suggested by the great masses
of pumice, at a distance from any volcanic eleva-
tion, which I found in the Cordillera of Pasto, between
Mamendoy and the Cerro del Pulpito, near the Eio
Mayo, 36 geographical miles north of the active vol-
cano of Pasto. Leopold von Buch has also called at-
tention to a similarly isolated outbreak of pumice,
described by Meyen, in Chili, east of Valparaiso, near
the village of Tollo, where it forms a hill about 300 feet
high. The volcano of Maypo, which in its upheaval
lifted up Jurassic strata, is two entire days' journeys from
this outbreak of pumice. (459) The Prussian Envoy at
Washington, Friedrich von Grerolt, to whom we owe
the first coloured geological map of Mexico, mentions
pumice-stone being obtained from beneath the surface
of the ground for building purposes, near Huichapa, 32
geographical miles south-east of Queretaro, at a distance
from any volcano. (46°) The geological investigator of

ON ITS EXTERIOR. VOLCANOES. 323

the Caucasus, Abich, is inclined, from his own observa-
tions, to regard the great eruption of pumice near the
village of Tschegem, on the northern side of the central
chain of Elbouruz, as an effect of action from fissures
much more ancient than the elevation of that conical
mountain, from which it is at a considerable distance,

If, according to the view which we have taken of vol-
canic activity, we regard the diminution of the earth's ori-
ginal temperature, by the radiation of its heat into space,
as the original cause, and the cooling and contraction
of the outer strata as the immediate effect, producing
fractures and foldings in the outer strata, in which
foldings some portions will be raised, and some portions
depressed (461), it will follow that the number of re-
cognisable volcanic frame-works (open cones and dome-
shaped elevations), upheaved over fissures, should be
regarded as the measure and evidence of the degree of
this activity in different regions. Such computations
have often been made in a very imperfect and unsatis-
factory manner, and sometimes mounts of eruption and
solfataras belonging to one and the same system, have
been enumerated as distinct volcanoes. The large
spaces in the interior of continents, which have hitherto
remained closed to all scientific examination, have not
been such great obstacles in the way of these inquiries
as may have been commonly supposed; as, generally
speaking, volcanoes are most often met with in islands,
and in regions near the coast. In a numerical inquiry,
which the state of our knowledge will not permit us to
render complete, much may still be gained by our
being able to assign an inferior limit, and to state, with
probably considerable approximation, the number of

r '2

324 REACTION OF THE INTERIOR OF THE EARTH

points at which3 within historic periods, the fluid in-
terior of the earth has remained in active intercourse
with the atmosphere. Such activity manifests itself,
most often simultaneously, by eruptions from volcanic
mountains, by increasing heat and inflammability of ther-
mal and naphtha springs, and in the increased extent of
areas of commotion: phsenomena which are all in intimate
interconnection and mutual dependence. (462) Leopold
von Buch has here again the great merit of having, in
writings appended to his physical description of the
Canary Islands, undertaken for the first time to em-
brace all the volcanic systems of the globe in a cosmical
view, while employing the well-grounded distinction of
central and linear systems of volcanoes. ' My own later,
and therefore more complete, enumeration, undertaken
on principles indicated above (pp. 243 and 265.) ex-
cluding, therefore, unopened domes, and mere cones of
eruption, gives, as a probable limitary mere minimum
number (nombre limite inferieur), a result which differs
considerably from any previously stated.

The question has been repeatedly raised, whether in
the parts of the earth's surface in which volcanoes are
most numerous, and the reaction of the interior upon
the solid crust of the earth most apparently active, the
molten interior may not actually be nearer to the sur-
face ? Whatever be the mode adopted for determining
the greatest mean thickness of the solid crust, whether
the purely mathematical one supplied by theoretical
astronomy (463), or the more simple one which is based
on the law of increase of heat with increasing depth, and
the melting point of rocks (364) ; still the problem pre-
sents a great number of, at present, indeterminate quan-

ON ITS EXTERIOR. VOLCANOES. 325

titles. We may cite as such the influence of enormous
pressure on the fusibility, the very different heat-con-
ducting power of different rocks, the remarkable dimi-
nution of the conducting power with great increase of
temperature treated of by Forbes, the unequal depth of
the bed of the ocean, and all the local accidents in the
connection and character of the fissures which lead down
to the fluid interior ! If the greater frequency of vol-
canoes and the more frequent and active intercourse
between the interior and the atmosphere in some regions
of the earth, is to be explained by the greater proximity
of the upper limit of the fluid interior to the surface,
this proximity may itself depend either on the relative
mean difference between the heights of the bed of the
sea and the continent, or on the unequal vertical depth
at which in different latitudes and longitudes the surface
of the molten fluid mass is situated. But where does
such a surface commence ? are there not intermediate
degrees between perfect rigidity and perfect fluidity?
Transitional states, which have often come into discussion
in controversies respecting the plasticity of some plu-
tonic and volcanic rocks which have been raised to the
surface, and respecting the movements of glaciers ? Such
intermediate states escape from mathematical considera-
tion as much as does the state of the so-called fluid
interior under an enormous pressure. If it be in itself
not altogether probable that heat should everywhere
continue to increase in arithmetical progression with
increasing depth, so also there may occur local inter-
vening disturbing causes, e. g. by subterranean basins,
or hollows, in the solid mass, which may be from time
to time partially filled from below, with fluid lava and

326 EEACTION OF THE INTERIOR OF THE EARTH

vapours resting upon it. (465) Such hollows were al-
ready made to play a part in the theory of decreasing
central heat by the immortal author of the " Protogaea,"
" Postremo credibile est contrahentem se refrigeratione
crustam bullas reliquisse, ingentes pro rei magnitudine,
id est, sub vastis fornicibus cavitates" (466) The more
improbable it is that the thickness of the already soli-
dified crust of the earth should be the same in all regions,
the more important is the consideration of the number
and geographical position of the volcanoes which are or
have been open within historic times. Such a view of
the geography of volcanoes can only be perfected by
often renewed attempts.

1. Europe.

Etna,

Volcano in the Li-

paris,
Stromboli,

Ischia,
Vesuvius,
Santorin,
Lemnos,

All belong to the great basin of the Mediterranean
Sea, but to its European, not its African shores : all these
seven volcanoes are or have been active within historic
times ; the burning mountain Mosychlos in the island of
Lemnos, which Homer terms the darling seat of He-
phaestos, was only destroyed and sunk beneath the waves
of the sea, together with the island of Chryse, by the
action of earthquake shocks, subsequently to the period
of the great Macedonian. (See Kosmos, Bd. i. S. 256
and 456, Anm. 9 ; Engl. ed. p. 230 and note 230 ;
Ukert, Geogr. der Griechen und Romer, Th. ii. Abth. i.
S. 198.) The great and, for almost 1900 years (from

ON ITS EXTERIOR. VOLCANOES. 327

186 B.C. to 1712), often repeated upheaving of the three
vents in the middle of the Gfulf of Santorin (partly
enclosed by Thera, Therasia, and Aspronisi), has borne
a striking similarity to the comparatively small pheno-
menon of the temporary formation of the islands called
Graham, Julia, and Ferdinandea, between Sciacca and
Pantellaria. On the peninsula of Methana, as has been
so often remarked (Kosmos, Bd. i. S. 453, Bd. iv. Amn.
86 to S. 273 ; Engl. ed. vol. i. note 230, and vol. iv. note
310), there are visible traces of volcanic eruptions in the
reddish brown trachyte which rises up out of the calca-
reous rock at Kaimenochari and Kaimeno (Curtius,
Pelop. Bd. ii. S. 439.)

Of prehistoric volcanoes, with fresh traces of out-
pourings of lava from craters, we have — proceeding from
north to south — the volcanoes of the Eifel (Mosenberg,
Greroldstein), as the most northern ; the great crater of
elevation, in which Schemnitz is situated ; Auvergne
(Chaine des Puys, or Mont Domes, le Cone du Cantal,
les Monts Dore) ; Vivarais, in which the ancient lavas
have broken forth out of gneiss (Coupe d'Aysac and the
Cone of Montpezat) ; Velay, eruptions of scoriae without
any lavas ; the Euganean Hills ; the Alban Hills, Eocca
Monfina and Vulturo, near Teano and Melfi ; the ex-
tinct volcanoes near Olot and Castell Follitt in Catalonia
(469) ; the group of islands called las Columbretes, near
the coast of Valencia (the larger crescent-shaped island,
Colubraria of the Eomans, has on it Montcolibre, in
39° 54' N. lat. according to Admiral Smyth, full of ob-
sidian and cellular trachyte) ; the Greek island Nisyros,
one of the Carpathian Sporades, quite circular in form,
having in its centre, at a height of 2270 feet according

328 REACTION OF THE INTERIOR OF THE EARTH

to Boss, a deep caldera, having an encircling ridge with
a strongly detonating solfatara, from which have radiated
streams of lava flowing into the sea, and now forming
small promontories ; this island furnished volcanic mill-
stones in the time of Strabo. (Boss, Reisen auf den Griech.
Inseln, Bd. ii. S. 69 and 72-78.) For the British Islands
I may here refer, on account of their antiquity, to the
remarkable effects of submarine volcanoes on the Lower
Silurian strata (Llandilo flags), inasmuch as volcanic
cellular fragments are imbedded in these strata ; and,
according to Sir Roderick Murchison's important obser-
vation, eruptive masses of trap even penetrate into the
Lower Silurian strata in the Corndon Hills (Shrop-
shire and Montgomeryshire) (468) ; the dyke phenomena
of the island of Arran ; and other places where there is
evidence of the introgression of volcanic activity, but no
traces of actual volcanoes have been found.

2. Islands in the Atlantic.

Volcano Esk on the island of Jan Mayen, ascended
by the meritorious Scoresby, and named by him after
his ship ; height rather less than 1600 feet. An open
but not burning summit crater ; basalt, rich in pyroxene
and trass.

South-west of the Esk volcano, near the north cape
of Egg Island, another volcano, which, in April 1818,
ejected ashes to a considerable height, at intervals of
four months.

Beerenberg, 6872 feet high, in the broad north-eastern
part of Jan Mayen, (71° 4' N. lat.) is not known to be a
volcano. (469)

ON ITS EXTERIOR. TOLC ANDES. 329

Volcanoes of Iceland : Oerafa, Hecla, Eauda-Kamba,
&c.

Volcano in the Island of Pico (47°), one of the
Azores : great eruption of lava from 1st of May to 5th
of June 1800.

Peak of Teneriffe.

Volcano of Fogo (471), one of the Cape de Verde
Islands.

Prehistoric volcanic activity. — This kind of volcanic
activity is less definitely connected with particular
centres in Iceland than in other places. If, with Sar-
torius von Waltershausen, we divide the volcanoes of
Iceland into two classes, one of such as have had only a
single eruption, and the other of such as have repeatedly
poured forth streams of lava from the same principal
fissure, then the first class will include Eauda-Kamba,
Skapta, Ellidavatan, south-east of Eeykiavik, &c. ;
while to the second class, which shows a more perma-
nent individuality, belong the two highest volcanoes of
Iceland, Oerafa (fully 6,400 feet high), and Snaefiall,
Hecla, &c. Snaefiall has not been in eruption within
the memory of man ; while Oerafa is known for its
dreadful eruptions in 1362 and 1727. ^Sart. von
Waltershausen, Phys. geogr. Skizze von Island S. 108
and 112.) In Madeira (472), the two highest moun-
tains : the conical Pico Euivo, 6059 feet high, and
Pico de Torres, which is only a little lower, and which
have their steep declivities covered with scoriaceous
lavas, cannot be regarded as the central points of the
former volcanic activity of the entire island, as at many
other points, and, especially towards the coasts, orifices
of eruption have been found, and even a large crater,

330 EEACTION OF THE INTERIOR OF THE EARTH

that of the Lagoa, near Machico. The lavas have been
thickened by confluence, and cannot be traced far as
separate currents. Remains of ancient dicotyledonous
vegetation, and of ferns, which have been accurately
examined by Charles Bunbury, are found buried in
raised beds of volcanic tufa and mud, and sometimes
covered by more recent basalts. Fernando de No-
ranha, 3° 50' S. lat., and 2° 27' east of Pernambuco, a
group of very small islands : rocks of phonolite, con-
taining hornblende ; no crater, but vein-fissures, filled
with trachyte and basaltic amygdaloid, traversing white
beds of tufa. (473) Island of Ascension, its highest
summit 2867 feet: basaltic lavas with more glassy
felspar interspersed than olivine, and in well defined
currents which can be traced to the trachytic cone of
eruption. Light-coloured trachyte, often in a tufa-like
state of disintegration, prevails in the interior, and in
the south-east part of the island. The masses of scoriae,
which have been thrown out from the Green Mountain,
contain imbedded angular fragments (474) in which there
are syenites and granites; these fragments remind us
of the lavas of Jorullo. To the west of the Green
Mountain there is a large open crater. Volcanic bombs,
partially hollow, sometimes of 10 or 11 inches diameter,
lie scattered about in great numbers, as do also large
masses of obsidian. St. Helena : the whole island is vol-
canic ; in the interior, there are more feldspathic beds of
lava ; towards the sea-shore, basaltic rock, traversed by
countless dikes, as in the Flagstaff Hill. Between
Diana's Peak and Nest Lodge, in the central mountain
range, there is " the mere wreck of a great crater " (475),
full of scoriae and cellular lava. The beds of lava are

ON ITS EXTEEIOE. YOLC ANDES. 331

not definitely circumscribed, and cannot, therefore, be
traced as proper lava-currents of small breadth. Tris-
tan da Cunha (37° 3' S. lat., 11° 27' W. long.), disco-
vered by the Portuguese in 1506, a small circular
island of 6 miles diameter, having for its centre a
conical mountain, which Capt. Denham describes as
being about 8,300 feet high, composed of volcanic rock.
(Petermann's Greogr. Mittheilungen, 1855, No. III. S.
84.) South-east of Tristan da Cunha, in 53° S. lat., is
the also volcanic Thompson's Island ; and intermediate
between them, in the same line, is (rough Island, also
called Diego Alvarez. Deception Island, a narrow,
ring-island, with a small opening (62° 55' S. lat.); and
Bridgman Island, belonging to the South Shetland
group, are both volcanic ; and amidst beds of ice, pumice,
black ashes, and obsidian, perpetually send forth hot
vapours. (Kendal in the Journal of the Greogr. Soc.
vol. i. 1831, p. 62.) In February 1842, flames were
seen to issue at once from thirteen points of the Ring
of Deception Island. (Dana in U.S. Explor. Exped. vol.
x. p. 548.) It is remarkable that while so many other
islands in the Atlantic are volcanic, neither the small
and quite flat Island of St. Paul (Penedo de S. Pedro),
one degree north of the equator (a very slightly lami-
nated greenstone-schist, passing into serpentine) (47G),
nor the Falklands (with their quartzose clay-slates), nor
South Georgia, nor Sandwich Land, appear to present
any volcanic rocks. On the other hand, a region of the
Atlantic about 0° 20' south of the equator, and 19° 40'
west long, from Greenwich, is considered to be the seat
of a submarine volcano (477). Krusenstern saw columns
of black smoke rise from the sea in this neighbourhood,

332 REACTION OF THE INTERIOR OF THE EARTH

(19th of May 1806), and the Asiatic Society in Calcutta,
in 1836, had laid before them ashes collected on two
different occasions at the same spot. According to very
careful inquiries by Daussy, it appears that on five
occasions, from 1747 to Krusenstern's voyage of circum-
navigation,— and on seven occasions, from 1806 to 1836,
— navigators within this volcanic region (as it is called in
the recent fine American chart of Lieut. Samuel Lee,
Track of the Surveying Brig Dolphin, 1854) have
remarked strange heavings of the sea, and shocks to
their vessels, which have been attributed to agitation of
the bottom of the sea by earthquake movements. Yet
very recently (Jan. 1852), in the Expedition of the
Brig Dolphin, which was instructed to take soundings
between the equator and 7 S. lat. and between 16 and
25 W. long., with reference to " Krusenstern's volcano,"
nothing remarkable was perceived, as had also been the
case in Wilkes's Exploring Expedition.

3. Africa.

The volcano Mongo-ma Leba in the Cameroon moun-
tains (N. lat. 4° 12') west of the mouth of the river of
the same name, in the Bight of Biafra, east of the
Delta of the Quorra or Niger, according to Captain
Allen, emitted a stream of lava in 1838. The line of
direction of four lofty volcanic islands, Annabona,
St. Thomas, Prince's Island, and Fernando Po, over a
S.S.W. N.N.E. fissure, points to the Cameroon mountain
which, according to Captain Owen and Lieutenant
Boteler, reaches the great height of about 13,000
feet. (478)

ON ITS EXTERIOR. VOLCANOES. 333

A volcano ? is believed to be a little to the west of
the snow-covered mountain Keenia, in Eastern Africa,
in about 1° 20' S. latitude, found in 1849, by the
missionary Krapf, near the sources of the Dana Eiver,
about 320 geographical miles north-west of the sea-
coast near Mombas. In a parallel nearly 2° more
southerly there is another snowy mountain, Kilimand-
jaro, discovered in 1847 by the missionary Eebmann,
scarcely 200 miles from the same coast. Eather more
to the west there is a third snowy mountain, Doengo
Engai, which was seen by Captain Short. The know-
ledge of the existence of these mountains has resulted
from courageous and perilous undertakings.

Proofs of prehistoric volcanic activity in the interior
of this great continent, but which has been so little
explored between the parallels of 7° N. and 12° S.
latitude, (those of Adamowa and the water dividing Lu-
balo mountains,) are presented (according to Euppell)
by the country round Lake Tzana, in the kingdom of
Grondar, as well as by the basaltic lavas, trachytes,
and obsidians of Shoa, according to Eochet d'Hericourt ;
the specimens brought home by the last-named traveller
have been examined by Dufrenoy, and found very
analogous to those of the Cantal and Mont Dore.
(Comptes Eendus, t. xxii. p. 806 — 810.) If the conical
mountain Koldghi in Kordofan has not shown itself to
be still active, yet the existence of black, porous,
vitrified rocks there has been confirmed. (479)

In Adamowa, south of the great Binue river, rise the
isolated mountain-masses of Bagele and Alantika,
which Dr. Barth, in his journey from Kuka to Jola,
judged by their cone and dome-shaped form to be

334 REACTION OF THE INTERIOR OF THE EARTH

probably trachytic mountains. Overweg, so early taken
away from the natural sciences, found in the district of
Grudscheba, examined by him west of Lake Tshad,
according to Petermann's notices derived from his
journals, columnar basaltic cones, rich in olivine, which
have broken through the beds of red argillaceous sand-
stone in some cases, and of quartzose granite in others.

A remarkable phenomenon is presented by the
paucity of still active volcanoes in this continent, whose
coasts, which are tolerably well known to us, present so
few indentations. May there be in the unknown
regions of Central Africa, especially south of the
equator, great basins of water analogous to Lake
Uniamesi (previously called N'yasse by Dr. Cooley), on
the shores of which volcanoes may rise like Demavend,
near the Caspian? Hitherto no reports of such have
reached us from any of the natives, who yet are great
travellers.

4. Asia.

a. THE WESTERN AND CENTRAL PORTION.

Volcano of Demavend (48°), burning, but according to
the accounts of Olivier, Morier, and Taylor Thompson
(1837), only moderately, and not smoking uninter-
ruptedly.

Volcano of Medina (eruption of lava in 1276).

Volcano Djebel el Tir (Taer or Tehr): an island-
mountain 895 feet high, in the Eed Sea, between
Loheia and Massaua.

Volcano Peschan: north of Kutsch in the great
chain of the Thian-schan, or Celestial mountains, in the
interior of Asia; eruptions of lava within a thoroughly

ON ITS EXTERIOK. VOLCANOES. 335

historical period, from the year 89 A.D. to the beginning
of the seventh century.

Volcano Ho-tscheu, also sometimes called, in the
very circumstantial Chinese descriptive geography,
volcano of Turfan: 120 geographical miles from the
great Solfatara of Urumtsi, near the east end of the
Thian-schan towards Hami, where much very fine fruit
is produced.

The volcano of Demavend, which rises to a height of
more than 19000 feet, is about thirty miles from the south
shore of the Caspian, in the province of Mazenderan ;
almost equidistant from Kescht and Asterabad, on the
chain of the Hmdu-Kho, which falls away rapidly
towards Herat and Meschid on the west. In another
work (Asie centrale, t. i. p. 124—129, t. iii. p. 433—435)
I have shown that it is probable that the Hindu-Kho,
from Chitral and Cafiristan, is a westerly continuation
of the great chain of the Kuen-lun which bounds Thibet
on the north, and at Tsunling crosses the Bolor
mountains which run north and south. Demavend
belongs to the Persian or Caspian Elburz or Elbourz;
employing that term as the name of a mountain-
system ; not to be confounded with the Caucasian
mountain summit, having a name of similar sound,
situated 7J° more to the north, and 10° more to the
west (now called Elburuz.) The word Elbourz is a
disfigurement of Albordj, the world-mountain, which is
connected with the very ancient cosmogony of the
Zend-people.

I have said that in taking a general view of the
direction of the mountain systems of Central Asia, we
may regard the volcano of Demavend, as being nearly

336 REACTION OF THE INTERIOR OF THE EARTH

the western extremity of the Kuen-lun ; we may also, I
think, regard with particular attention, as being at
their easternmost extremity, another igneous pheno-
menon, whose existence was first made known by me
(Asie centrale, t. ii. p. 427 and 483). In the important
researches which my honoured friend and colleague in
the Institute, Stanislas Julien, has undertaken, at my re-
quest, for the purpose of obtaining from the rich geogra-
phical sources of the old Chinese literature, information
respecting the Bolor, the Kuen-lun, and the " starry-
sea," he found in the great dictionary edited by the
Emperor Yongtsching in the beginning of the eighteenth
century, a description of the " perpetual flame " which
issues forth from a cave in the hill Schinkhieu on the
slope of the eastern part of the Kuen-lun. This pheno-
menon, the light of which is seen from a great distance,
cannot well be called a volcano : it appears rather to be
analogous to the Chimera in Lycia, famous of old
among the Greeks. It is a fountain of fire, a spring of
gas, which the volcanic activity of the interior of the
earth keeps always lighted (Kosmos, Bd. iv. S. 296,
note 51 ; Eng. ed. p. 252 and note 375.)

Arabian writers say, but generally without assigning
any particular year, that, in the middle ages, erup-
tions of lava took place at particular points on the
south-west coast of Arabia, in the islands of Zobayr
in the straits of Bab-el-Mandeb and at Aden (Wellsted,
Travels in Arabia, vol. ii. p. 466 — 468), in Hadhramaut,
in the straits of Ormuz, and in the western part of the
Persian gulf; — always upon ground which had been
the seat of volcanic activity since prehistoric times.
The epoch of the eruption of a volcano at Medina

ON ITS EXTERIOR. VOLCANOES. 337

itself, 12^ degrees north of the straits of Bab-el-Mandeb,
has been found by Burckhardt in Samhudy's chronicle
of that celebrated town. It corresponds to the
2nd of November 1276. But we also learn from
Abulmahasen that an igneous eruption had taken place
there in 1254, or twenty-two years earlier. (Compare
Kosmos, Bd. i. S. 256, Eng. p. 234.) The island volcano,
Djebel Tair, which was already recognised by Vincent
as the "extinct once-burning island" of the Periplus
Maris Erythrsei, is still active and sends forth smoke,
according to Botta, and according to information
collected by Ehrenberg and Russegger (Reisen in
Europa, Asien und Afrika, Bd. ii. Th. i. 1843, S. 54).
Respecting the whole region round Bab-el-Mandeb,
with the basaltic island of Perim; the crater-like
encircling ridge within which the town of Aden is
situated ; the island of Seerah with streams of obsidian
which are covered with pumice ; and respecting the
groups of islands of Zobayr and Farsan (the volcanic
character of the latter was discovered by Ehrenberg in
1825), see the fine investigations of Ritter in his
Erdkunde von Asien, Bd. viii. Abth. 1. S. 664 — 707,
889—891, and 1021—1034.

The volcanic range of the Thian-schan (Asie centrale,
t. i. p. 201 — 203, t. ii. p. 7 — 61), a mountain system
extending from west to east across Central Asia from
the Altai to the Kuen-lun, was at one time an especial
object of study to me, when to the little which Abel
Remusat had obtained concerning it from the Japanese
Encyclopedia, I was able to add the more important
fragmentary notices discovered by Klaproth, Neumann,
and Stanislas Julien (Asie centrale, t. ii. p. 39 — 50 and

VOL. iv z

338 HE ACTION OF THE INTERIOR OF THE EARTH

335 to 364). The Thian-schan chain is more than
eight times the length of the Pyrenees, if passing
beyond the north and south chain of the Kusyurt-Bolor,
which is traversed by it, we include the Asferah, which
extends westward to the meridian of Samarcand, and in
which, as in the Thian-schan, Ibn Haukal and Ibn al
Vardi describe fountains of fire and luminous ? clefts
emitting sal-ammoniac. (See on Mount Botom, Asie
centrale, t. ii. p. 16 — 20.) In the history of the dynasty
of Thang it is expressly said, that, on one of the declivi-
ties of the Peschan, which perpetually sends forth fire
and smoke, the stones burn, melt, and flow for several
li " like liquid fat ; the soft mass hardens as it cools."
It is scarcely possible to indicate a current of lava
more distinctly and characteristically. In the 49th book
of the voluminous Geography of the Chinese Empire,
which was printed at the expense of the government in
Pekin from 1789 to 1804, the fire-mountains of the
Thian-schan are described as still active. Their situa-
tion is so central that they are about equidistant (1520
geographical miles) from the nearest part of the Icy
Sea and the mouths of the Indus and the Ganges;
1020 miles from the sea of Aral, and 172 and 208 miles
from the salt lakes of Issikal and Balkasch. The flames
which rise from the mountains of Turfan (Hotscheu)
have also been mentioned by pilgrims to Mecca, who
were officially questioned in 1835 at Bombay. (Journal
of the Asiatic Soc. of Bengal, vol. iv. 1835, p. 657 — 664.)
When will the volcanoes of Peschan and Turfan, Barkul
and Kami, be visited by a scientifically qualified tra-
veller, who might proceed from Gouldja on the Hi,
which can be so easily reached ?

ON ITS EXTERIOR. VOLCANOES. 339

The now better-known situation of the volcanic chain
of the Thian-schan has very naturally led to the question,
whether the mythical land of Gog and Magog, where
perpetual fires are said to bum from the bottom of the
Eiver el Macher, may not be connected with the erup-
tions of Peschan or the volcano of Turfan ? This Oriental
myth, which belonged originally to the west side of the
Caspian, the Pylse Albanise, near Derbend, has migrated,
as is so often the case with fables, and has travelled far
to the eastward. Edrisi makes Salam el Terdjeman,
interpreter to one of the Abasside caliphs, travel from
Bagdad, in the first half of the ninth century, towards
the Land of Darkness. Passing through the Steppe of
the Baschkirs, he arrives at the snow-covered mountain
Coccai'a, which is surrounded by the great wall of Magog
(Madjoudj). Amedee Jaubert, to whom we owe im-
portant supplements to our knowledge of Nubian geo-
graphy, has shown that the fires which burn on the
declivity of the Coccai'a are not volcanic (Asie centrale,
t. ii. p. 99). Edrisi places further to the south the
Lake Tehama. I think I have shown the probability
of Tehama being the great lake Balkasch, into which the
Hi falls, and which is only 180 miles to the south. A
century and a half after Edrisi, Marpo Polo transferred
the wall of Magog to the mountain of In-schan, east of
the high plain of Grobi, towards the river Hoang-ho and
the Chinese Wall, of which, as well as of the use of tea,
the celebrated Venetian traveller, strangely enough, does
not speak. The In-schan, the boundary of the domi-
nions of Prester John, may be regarded as the eastern
prolongation of the Thian-schan (Asie centr. t. ii. p. 92
-104).

r- O

340 REACTION OF THE INTERIOR OF THE EARTH

The two once lava-erupting conical mountains, the
volcano Peschan and the Hotscheu of Turfan, were long
erroneously regarded as forming one isolated volcanic
group. They are almost 420 geographical miles apart,
and are separated by a considerable mountain mass,
Bogdo-Oola, covered by perpetual ice and snow. I
think I have shown that, as in the Caucasus, the vol-
canic activity north and south of the long chain of the
Thian-schan is in close geological connection with the
limits of the circle of earthquake commotion, the hot
springs, solfataras, fissures emitting sal-ammoniac, and
beds of rock-salt.

As, according to my often expressed view, now fully
participated in by the observer who is so thoroughly
acquainted with the Caucasian mountain-system, Abich,
the Caucasus itself is only to be regarded as belonging
to the continuation of the fissure of the volcanic Thian-
schan and Asferah, beyond the great Aralo-Caspian
depression (481), we should here mention, by the side of
the phenomena of the Thian-schan, as belonging to pre-
historic epochs, the four extinct volcanoes; Elburuz
18,493 feet high, Ararat 17,112 feet, Kasbegk 16,532
feet, and Savalan 15,759 feet. (482) In elevation these
volcanoes are intermediate between Cotopaxi and Mont
Blanc. The Great Ararat (Agri-dagh), first ascended
on the 27th September 1829 by Friedrich von Parrot,
repeatedly in 1844 and 1845 by Abich, and lastly in
1850 by Colonel Chodzko, is dome-shaped like Chimbo-
razo, and has two small elevations at the edge of the
summit, but no summit-crater. The greatest, and pro-
bably the most recent, prehistoric lava eruptions of
Ararat have all broken forth below the limit of per-

ON ITS EXTERIOR. VOLCANOES. 341

petual snow ; the erupted substances are of two kinds ;
part being trachytic with glassy felspar, and interspersed
pyrites very susceptible of weathering ; and part dole-
Titic, consisting chiefly of labradorite and augite, like the
lavas of Etna ; Abich regards the latter as, in this case,
the more recent. The places from which the streams
of lava have flowed (all of which, as already stated, are
below the snow limit) are often marked by cones of
eruption, and small craters surrounded by scoriaB (for
example in the large grassy plain of Kip-Grhioll on the
north-western declivity). Although the deep valley of
St. James (a ravine which runs up to the summit of
Ararat, and gives to its form a peculiar character, even
when seen at a very great distance) presents much
similarity to the Val del Bove of Etna, and renders
visible the very interior structure of the upheaved dome,
yet there is a striking difference, inasmuch as in St.
James's Valley trachytic rock is found in mass, without
lava currents, or beds of scoriaB or of rapilli. (483) The
Greater and the Lesser Ararat, the first of which is,
according to the excellent geodesical operations of Vasili
Fedorow, 3' 4" to the north and 6' 42" to the west of
the latter, rise on the southern edge of the great plain
through which the Araxes rolls in a wide sweep. They
both stand on the same ellipticaily shaped volcanic
plateau, whose major axis is directed from south-east
to north-west. Kasbegk and Tschegem are also without
summit-craters, although the former has sent forth
great eruptions in a northerly direction (towards Wladi-
caucas). The greatest of all these extinct volcanoes,
the trachytic cone of Elburuz, whi^h has been up-
heaved from the schistose, talcose, dioritic, and graniti-

3^2 REACTION OF THE INTERIOR OF THE EARTH

ferous mountains of the valley of the Backsan Elver,
has a crater-lake. Similar crater-lakes are found in the
rough highland of Kely, and lava currents can be seen
to have flowed from them between cones of eruption.
Here, as in the Cordilleras of Quito, the basalts are
widely separated from the trachyte systems ; they only
begin twenty-four and thirty-two miles south of the
Elburuz chain, and of Tschegem, on the upper valley
of the Phasis or Rhion.

|8. THE NORTH- EASTERN PART or ASIA.
(Peninsula of Kamtschatka.)

The Peninsula of Kamtschatka, from Cape Lopatka
(according to Krusenstern in 51° 3 N. lat.) northward
to Cape Ukinsk, belongs to the same category as the
Island of Java, Chili, and Central America, these being
the regions on the earth's surface where the greatest
number of volcanoes, and, moreover, the greatest num-
ber of still active volcanoes are congregated together
on a comparatively small area. In Kamtschatka four-
teen are enumerated in a length of 420 geographical
miles. For Central America, I find, in 680 such
miles (from the volcano of Soconusco to Turrialva
in Costa Rica), twenty-nine volcanoes, of which eighteen
are burning ; for Peru and Bolivia in 420 miles (from
the volcano Chacani to that of San Pedro de Atacama),
fourteen volcanoes, of which three are at present active ;
for Chili, in 960 miles (from Volcan de Coquimbo to
V. de San Clemente), twenty-four volcanoes, of which
thirteen are known to have been active within historic

ON ITS EXTERIOR. VOLCANOES. 343

times. Our knowledge of the volcanoes of Kamtschatka,
in respect to form, and exactly determined geographical
position and elevation, has received admirable enlarge-
ment in modern times, through the exertions of Krusen-
stern, Horner, Hofmann, Lenz, Liitke, Postels, Beechey,
and above all Adolph Erman. The peninsula is inter-
sected lengthways by two parallel chains, in the eastern-
most of which the volcanoes are chiefly or wholly si-
tuated. The highest of these attain elevations of from
10,500 to 14,800 French feet (11,190 to 15,773 Eng-
lish feet). They succeed each other in order, as fol-
lows : proceeding from south to north.

The Opalinski Volcano (Peak Koscheleff of Admiral
Krusenstern), lat. 51° 21': according to Captain Chwos-
tow, its elevation is almost equal to that of the Peak of
Teneriffe, and it was exceedingly active at the end of
the 18th century.

Hodutka Sopka (51° 35' lat.). Between this mountain
and the last there is an unnamed volcanic cone (51° 32'),
which, however, like Hodutka, appears, according to
Postets, to be extinct.

Poworotnaja Sopka (52° 22' lat.), according to Cap-
tain Beechey 7931 feet high (Erman's Reise, Bd. iii. S.
253 ; Leop. von Buch, lies Can. p. 447).

Assatschinskaja Sopka (52° 2' N. lat.) ; great eruptions
of ashes, especially in 1828.

Wiljutschinsk Volcano (lat. 52° 52'): according to
Captain Beechey 7373 feet high, according to Admiral
Liitke 6746 feet; only twenty geographical miles from
the port of St. Peter and St. Paul, beyond the bay of
Torinsk.

Awatschinskaja, or Grorelaja Sopka (lat. 53° 17'),

344 REACTION OF THE INTERIOR OF THE EARTH

height, according to Erman, 8910 feet; first ascended
in the expedition of La Perouse, in 1787, by Mongez and
Bernizet; and subsequently by my dear friend and
Siberian travelling companion, Ernst Hofmann (July
1824, in Kotzebue's Voyage of Circumnavigation); by
Postels and Lenz in Admiral Liitke's expedition in 1828 ;
and by Erman in September 1829. Erman made the
important geological observation that the trachyte at its
upheaval had broken through schist and graywacke
(a Silurian rock). This still smoking volcano had a
dreadful eruption in October 1837, having previously
had a much slighter one in April 1828. Postels in
Liitke's Voyage, t. iii. p. 67 — 84 : and Erman, Eeise,
hist. Bericht, Bd. iii. S. 494, and 534—540.

Quite near to the Awatscha Volcano (see note 345 of
the present volume) there is Koriatskaja or Strjeloschnaja
Sopka (lat. 53° 19'), height 11,210 feet according to
Liitke, t. iii. p. 84 ; rich in obsidian, which the people
of the country continued, so late as the last century, to
use as the Mexicans, and, at a very early period of their
history, the Greeks had done, for arrow-points.

Jupanowa Sopka: lat., according to Erman's deter-
mination (Eeise, Bd. iii. S. 469), 53° 32'. The summit
is rather flattish ; and Erman's remark upon it is, " that
this Sopka, from the smoke which it emits, and the
subterranean noises which are heard, has always been
compared to the great volcano of Schiwelutsch, and
reckoned among undoubted burning mountains." Its
height was measured from the sea by Liitke, who made
it 9055 feet.

Kronotskaja Sopka, 10,609 feet, on the lake of the
same name, lat. 54° 8' ; a smoking crater at the summit

ON ITS EXTERIOR. VOLCANOES. 345

of a very pointed conical mountain. (Ltitke, Voyage,
t. iii. p. 85.)

Schiwelutsch, twenty miles southeast of Jelowka, was
almost unknown before it was visited and examined by
Erman, from whom we now possess an excellent account
of it (Reise, Bd. iii. S. 261—317, and Phys. Beob. Bd.
i. S. 400—403) : its northern point is in lat. 56° 40'
and its height 10,544 feet ; southern point lat. 56° 39',
height 8793 feet. When Erman ascended Schiwelutsch
in September 1829, he found it smoking strongly. Great
eruptions took place in 1739, and between 1790 and 1810,
the latter having been not of flowing liquid lava, but
of loose volcanic rocks. According to C. von Dittmar,
the northernmost summit fell in on the night of 17-18
February 1854, after which there followed an eruption
of long continuance, which, possibly, has not yet ceased,
accompanied by actual streams of lava.

Tolbatschinskaja Sopka : emits great outbursts of
smoke, and has, in former times, often varied the place
of the eruption-orifices, from which ashes have been
ejected; according to Erman, lat. 55° 51', and height
8313 feet.

Uschinskaja Sopka : nearly connected with the Kliu-
tschewsk volcano, lat. 56° 0', height (11,000 French)
11,723 English feet (von Buch, Can. S. 452; Land-
grebe, Vulkane, Bd. i. S. 375.)

Kliutschewskaja Sopka (lat. 56° 4') : the highest and
most active of all the volcanoes in this peninsula, has
been thoroughly examined by Erman, both geologically
and hypsometrically. According to the accounts of Kras-
chenikoff, it had great igneous eruptions from 1726 to
1731, and also in 1767 and 1795. In 1829, Erman, in his

346 REACTION OF THE INTERIOR OF THE EARTH

dangerous ascent of the volcano, was himself an eyewit-
ness, on the 1 1th of September, of the eruption of glowing
stones, ashes, and vapours from the summit, while, far be-
low, a considerable stream of lava poured itself forth from
a fissure on the western declivity. Here, also, the lava is
rich in obsidian. According to Erman (Beob. Bd. i. S.
400 — 403, and 419), the latitude of the volcano is 56° 4',
and its height, in September 1829, was very exactly
1 5,763 feet. On the other hand, in August 1828, Admiral
Liitke, by angles of altitude taken from the sea at the
distance of forty nautical miles, made it 16,500 feet
(16,498). (Voyage, t. iii. p. 86 ; Landgrebe, Vulkane,
Bd. i. S. 375 — 386.) This measurement, and the com-
parison of the excellent outline drawings of Baron von
Kittliz, who accompanied Liitke's expedition in the Senia-
vin, with what Erman himself observed in September
1829, led the latter to believe that, in this short inter-
val of thirteen months, great changes had taken place
in the shape and height of the summit. He says
(Reise, Bd. iii. S. 359), " we can scarcely be much in
error if we assume the height of the summit in August
1828, to have been 250 French feet (266 English)
higher than in September 1829 ; making the height, at
the earlier epoch (in round numbers) 16,030 feet."
On Vesuvius, taking as my basis Saussure's barometric
measurement of the Rocca del Palo (the highest north-
ern edge of the crater) in 1773, I found, by my own
measurements, that in 1805 — therefore, in an interval
of thirty-two years — this edge had sunk thirty-eight
and one third feet; but that, from 1773 to 1822, or in
forty-nine years, it had gained (apparently ?) 102 feet.
(Ansichten der Natur, 1849, Bd. ii. S. 290.) In 1822,

ON ITS EXTERIOR. VOLCANOES. 347

Monticelli and Covelli found for the Rocca del Palo
3990 feet, and I found nearly 4022 feet. I assumed
as the most satisfactory final result for that time 3997
feet. In the spring of 1855, therefore, thirty-three years
later, the fine barometrical measurements of the Olmutz
astronomer, Julius Schmidt, again gave 3990 feet. (Neue
Bestimm. am Vesuv., 1856, S. 1, 16, and 33.) How
much of these differences is due to error of observa-
tion, barometric formulae, and casual circumstances inci-
dental to the method? Such investigations could be
carried on more extensively and securely, if, instead of
often repeated complete trigonometrical operations, or,
in the case of accessible summits, of the more easily
applied, but less satisfactory, barometric method, it
were arranged simply to take, at desirable intervals,
say twenty-five or fifty years, angles of altitude of the
crater margin from some definite and easily identified
spot, but making those angles exact to fractions of
seconds. On account of the influence of terrestrial
refraction, I would advise that, at each normal epoch,
the mean of observations at several different hours and
on three different days should be obtained. In order
to have not merely the general result of increase or
diminution, but also the absolute amount of change
•expressed in feet, it would suffice to determine, once for
all, the distance of the point of observation. What a
rich source of knowledge, founded on experience extend-
ing over more than a century, respecting the colossal
volcanoes of Quito, would now be open to us, if, in addi-
tion to the sufficiently exact measurements left to us by
Bouguer and La Condamine, those distinguished men
had furnished us with the knowledge of exact spots, per-

348 EEACTION OF THE INTERIOR OF THE EARTH

manently marked, as the sites from which the angles of
altitude of the respective summits had been taken ?
According to C. von Dittmar, after its eruption in 1841,
Kliutschewsk was completely at rest until its reawaken-
ing in 1853, when it sent forth lava; but the falling
in of the summit interrupted its renewed activity.
(Bulletin de la Classe physico-mathem. de 1'Acad. des
Sc. de St. Petersbourg, t. xiv. 1856, p. 246.)

There are yet four other volcanoes, mentioned in part
by Admiral Lutke and in part by Postels, — Apalsk,
which is still smoking, to the south-east of the village
of Bolscheretski ; Schischapinskaja Sopka (lat. 55° 11');
the conical Krestowsk (lat. 5 6° 4'), near the Kluitschewsk
group; and Uschkowsk; — which I have not included
in the preceding list, for want of more exact determina-
tions. The central mountains of the peninsula, espe-
cially in the Baidar plains, in lat. 57° 20', on the east
of Sedanka, present (as if the plain were " the floor of
a very ancient crater of about a league in diameter ")
the geologically remarkable phenomenon of lava and
scoriae, which have been poured forth from a brick-red
volcanic rock, full of bubbles, which has itself broken
forth through fissures in the earth, all at a considerable
distance from any upheaved conical volcanoes of regu-
lar structure (Erman, Reise, Bd. iii. S. 221, 228, and
273 ; von Buch, lies Canaries, p. 454). There is here
a striking analogy to the circumstances which I have
developed in detail above (p. 304), respecting the Mal-
pais and problematical fields of rocky fragments in the
Mexican highlands.

ON ITS EXTEKIOK. YOLCANOES. 349

5. East Asiatic Islands.

From Torres Strait, which in 10° S. lat. divides New
Guinea from Australia, and from the smoking volcanoes
of Flores to the north-easternmost Aleutian Islands in 55°
N. lat., there extends an island world, which is for the
greater part volcanic, and which, when regarded in a ge-
neral geological point of view by reason of its genetical
connection, cannot very easily be separated into single
groups, and of which the area widens considerably
towards the south. Beginning in the north and pro-
ceeding from the American peninsula Alaska, we see
the bow-shaped curvature of the range of Aleutian
Islands (484), uniting, through the island of Attu which
is near Copper and Bering's Islands, the old and the new
continent ; and, as it were, enclosing Bering's sea by a
boundary towards the south. If we, further, proceed
southwards from Cape Lopatka, the extremity of the
peninsula of Kamtschatka, we have first the Kurile
Islands, forming the eastern boundary of the sea of
Saghalin or Ocho-tsk, which La Perouse rendered cele-
brated; next Jezo, which perhaps was once connected
with the southern point of the island of Krafto (485)
(Saghalin or Tschoka) ; and lastly, beyond the narrow
straits of Tsugar, the three islands which form the
Japanese empire (Nippon, Sikok, and Kiusiu, lying,
according to Siebold's excellent map, between 41° 32'
and 30° 18' N. latitude). From the volcano of Kliu-
tschewsk, the northernmost one on the east coast of the
Kamtschatkan peninsula, to the southernmost Japanese
island-volcano of Iwoga-Sima in the strait of Van Die-

350 REACTION OF THE INTERIOR OF THE EARTH

men, explored by Krusenstern, the direction in which
the igneous activity manifests itself from the fissured
crust of the globe is exactly from north-east to south-
west. This direction is maintained through the island
Jakuno-Sima, on which a conical mountain rises to the
height of 5840 feet, separating the two straits of Van
Diem en and Colnet ; through Siebold's Linschote Archi-
pelago ; through Captain's Basil Hall's Sulphur Island
(Lung-Huang-Schan) ; and through the small groups of
Lieu-Khiew and Madjiko-Sima, which latter approaches
within 92 geographical miles of the great island off the
Chinese coast, Formosa (Thay-wan).

Here, at or near Formosa, in 25° and 26° N. lat., we
may recognise the important point at which, instead of
the north-east and south-west lines of elevation, those of
a north and south direction commence, and prevail
almost to the parallels of 5° or 6° S. lat. These K— S.
lines may be traced in Formosa and the Philippines
(Luzon and Mindanao), through fully twenty degrees of
latitude ; cutting off, as it were, the coasts on either or on
both sides ; as for example, the east coast of the large
island of Borneo, which is connected with Mindanao by
the Solo Archipelago, and with Mindoro by the long
narrow island of Palawan ; and the western portion of
Celebes with its strangely varied outline, and Gilolo ;
and (which is especially remarkable) appearing in the
longitudinal fissure, over which, 1400 geographical miles
east of the group of the Philippines and in the same
latitude, the series of volcanic and coral-islands of the
Marianas or Ladrones has been elevated. Its general
direction is N. 10° E. (486)

As we have marked the parallel of the coal-containing

ON ITS EXTERIOE. VOLCANOES. 351

island of Formosa as the turning point where the Kurile
N.E.S. — W. direction is followed by a N. — S. one, so we
may, in like manner, point to a new fissure-system as
commencing to the south of Celebes and Borneo whose
southern coast is cut east and west. The greater and
lesser Sunda islands from Timor-Laut to West-Bali
follow for the most part the mean parallel of 8° S. lat.,
through eighteen degrees of longitude. In the west
part of Java the middle axis already turns rather more
towards the north, running almost E.S.E. — W.N.W., but
from the Straits of Sunda to the southernmost of the
Nicobars the direction is S.E. — N.'W. The entire vol-
canic fissure of elevation (E.— W. and S.E.— KW.)
has, according to this, an extent of about 2700 geogra-
phical miles, or eleven times the length of the Pyre-
nees ; of the whole distance, if we disregard the slight
deviation in Java towards the north, 1620 miles belong
to the east and west, and 1080 to the south-east and
north-west direction.

In this manner, geological considerations on form and
arrangement conduct us uninterruptedly through the
islands of the eastern coast of Asia, over the enormous
space of 68 degrees of latitude, from the Aleutian
islands and the northern sea of Bering to the Moluccas
and the greater and lesser Sunda Isles. It is especially
in the zone comprised between 5° N. and 10° S. latitude
that the most abundant variety of configuration of land
has been developed. The directions of the lines of out-
burst or elevation of the larger parts are most frequently
seen repeated in a remarkable manner in neighbouring
small ones. Thus, near the south coast of Sumatra, we
have a long row of islands parallel to it. We may

3J2 EEACTION OF THE INTERIOR OF THE EARTH

remark in the small phenomenon of veins of ore the same
kind of agreement as in the great one of lines of moun-
tains, or of islands, extending over entire continents or
extensive regions of the globe. Secondary veins having
the same "strike" as the principal one, and accompany-
ing or secondary chains of mountains (" chaines accom-
pagnantes"), are often found at considerable distances;
they point to the same causes and to the same directions
of the form-giving agency in the folding or wrinkling of
the earth's crust. The conflict of forces in the simulta-
neous opening of fissures in opposite directions appears
to produce now and then extraordinary juxtapositions of
form ; as in the complicated outlines of Celebes and
Gilolo.

Having thus indicated the inherent geological con-
nection of the eastern and southern Asiatic islands, we
will in order not to depart from the old, somewhat
arbitrary, geographical divisions and nomenclature, take
the southern limit of the East-Asiatic chain of islands
at the point where near Formosa the direction turns or
passes, as already stated, from N.E. — S.W. to N. — S.,
in the 24th degree of north latitude. The enumeration
is again made from north to south ; beginning with the
easternmost, rather American than Asiatic, Aleutian
Islands.

The Aleutian islands, rich in volcanoes, comprise,
proceeding from east to west ; the Fox islands, among
which are the largest of the entire series Unimak, Una-
laschka,and Umnak; the Andrejanowski islands, amongst
which the most celebrated are Atcha with three smok-
ing volcanoes, and the island of the great volcano
of Tanaga, which has been drawn by Sauer; the Rat

ON ITS EXTERIOR. VOLCANOES. 353

islands and the Blynie islands, which stand somewhat
apart ; amongst these Attu forms, as we have already
said, the transitional link with the Asiatic " Commandeur
group" (Copper and Bering's Islands). There appears to
be but little foundation for the often repeated statement,
that the line of continental volcanoes in the peninsula of
Kamtschatka, directed from N.KE.to S.S.W., only begins
at the place where the volcanic elevation fissure of the
Aleutian islands, passing under the sea, intersects the
peninsula ; thus representing the Aleutian fissure as a
channel of conduction. According to Liitke's map . of
Bering's Sea, the island of Attu, the westernmost extre-
mity of the Aleutian Islands, is in 52° 46' lat., and the
non-volcanic Copper and Bering's islands are in 54° 30'
and 55° 20', while the line of Kamtschatkan volcanoes
commences in 56° 40' with the great volcano of Schi-
welutsch, west of Cape Stolbowoy. The direction of the
eruptive fissures is also very different, almost opposite.
The loftiest of the Aleutian volcanoes, that in Unimak,
is, according to Liitke, 8076 feet high. Near to the
north point of Umnak, in May 1796, the island of
Agaschagokh (or Sanctus Johannes Theologus) was up-
heaved from the sea under very remarkable circum-
stances, which are extremely well described in Kotzebue's
Voyage of Discovery (Bd. ii. S. 106); and it continued
burning for almost eight years. According to an
account communicated by Krusenstern, it was, in 1819,
almost 16 geographical miles in circumference, and still
upwards of 2200 feet high. On the island of Unalaschka,
the relations assigned by the ingenious Chamisso of the
trachyte containing much hornblende of the volcano Ma-
tuschkin (5474 feet high) to the black porphyry (?) and

VOL. IV. A A

354 REACTION OF THE INTERIOR OF THE EARTH

adjacent granite, deserve to be examined by an observer
well acquainted with modern geological science, and the
mineralogical composition of the rocks to be accurately
investigated. Of the two islands of the Pribytow group,
which stand a little apart by themselves iD Bering's sea,
St. Paul's is entirely volcanic with much lava and
pumice ; although, on the other hand, St. Greorge's only
contains granite and gneiss.

According to the most complete accounts which we at
present possess, the range of the Aleutian Islands, 960
miles long, appears to contain upwards of 34 volcanoes,
most of which have been active within recent historic
times. Thus, we have here (within 54° and 60° N. lat,
and 160° to 196° W. long.) a strip of the whole sea bot-
tom between two great continents in a state of constantly
alternating, formative and destructive activity. Here,
as in the Azores, in the course of thousands of years
many islands may have been elevated to near the surface
of the sea, yet without having actually risen above it,
and many may long since have appeared and have sub-
sided again, either wholly or partially unobserved. For
the migration and intermixture of races and people, the
Aleutian Islands offer a route 13 or 14 degrees south
of that by Bering Strait, by which the Tschuktsches
seem to have passed from America to Asia, and even to
beyond the Anadyr River.

The range of the Kurile Islands, extending from the
extreme point of Kamtschatka to Cape Broughton
(the north-easternmost promontory of Jezo), a length of
720 geographical miles, has 8 or 10 volcanoes, most
of which are still burning. The northernmost of these
in the island Alaid is known for its great eruptions

ON ITS EXTERIOR. VOLCANOES. 355

in 1770 and 1793, and its height would appear to be
well-deserving of exact measurement, being estimated
at between 12,000 and 15,000 feet. The far lower peak
Sarytschew (4505 feet, according to Horner) in Mataua,
and the southernmost Japanese Kuriles, Urup, Jetorop
and Kunasiri have also shown themselves to be still very
active volcanoes.

Next in our volcanic series follow Jezo and the three
great Islands of Japan, upon which the celebrated tra-
veller Herr von Siebold has kindly communicated to me
a very great and important notice, for employment in
the Cosmos. It will rectify and complete the accounts
taken from the great Japanese Encyclopedia in my
Fragmens de Greologie et de Climatologie Asiatiques
(t. 1, p. 217—234), and Asie centrale (t. 11, p. 540
— 552).

The large, and in its northern portion very square-
shaped, island of Jezo (41 J° to 45^° lat), divided by the
Sangar or Tsugar Strait from Nippon, and by the Strait
of La Perouse from Krafto ("or Karafuto), bounds by its
north-easternmost cape the range of the Kuriles ; but not
far from the north-westernmost point, Cape Romanzow,
which advances a degree and a half further to the north
on the Strait of La Perouse, there is in 45° 11' lat. the
volcanic Pic de Langle (5350 feet high), on the small
island of Eisiri. Jezo itself appears to be intersected
by a range of volcanoes running from Broughton's
southern Volcano bay towards the northern cape ; and
this is the more deserving of attention, because on
the narrow island of- Krafto, which may almost be
regarded as a continuation of Jezo, the naturalist of La
Perouse's expedition found in the Baie de Castries red

A A 2

356 REACTION OF THE INTERIOR OF THE EARTH

porous lavas and fields of scoriae. In Jezo itself, Siebold
counts seventeen conical mountains, of which the
greater part appear to be extinct volcanoes. Kiaka,
called by the Japanese Usuga-Take, or the " Mortar-
mountain," from a deeply sunken crater, and Kajo-hori,
both appear to be still burning. (Commodore Perry
saw two volcanoes at the harbour of Endermo, lat. 42°
17'.) The lofty range (Krusenstern's conical mount
Pallas) is in the middle of the Island of Jezo, in about
44° lat., and about E.N.E. of Strogonow Bay.

" The historians of Japan mention only six volcanoes
as having been active after or before our era, two in the
island of Nippon, and four in that of Kiusiu. The
volcanoes of Kiusiu, the island nearest to the Corean
peninsula, are, taking them according to their geogra-
phical position from south to north: (1) Mitake volcano,
on the little island Sayura-Sima, in the bay of Kagosima,
which is open towards- the south, in the province of
Satsuma (lat. 31° 33', long. 130° 43' E.); (2) Volcano
Kirisima, in the district of Naka, and in the province of
Fiuga (lat. 31°45/); (3) Asojama, in the district Aso,
province Figo (lat. 32° 45' ) ; (4) Wunzen, in the pen-
insula of Simabora, in the district Takaku (lat. 32°
44'). The height of the last-named volcano amounts,
according to barometric measurement, to only 4110 feet,
only about a hundred feet higher than Vesuvius (Eocca
del Palo). Its most violent eruption, historically known
to us, was that of February 1793. The volcanoes
Wunzen and Asojama are both east-south-east of Nan-
gasaki.

" The volcanoes of the great island of Nippon are,
again beginning with the south: — (1) Fusijama, scarcely

ON ITS EXTERIOR. VOLCANOES. 357

sixteen geographical miles from the south coast, in the
district of Fusi, province Suruga (lat. 35° 18', long.
138° 37' E.). Its height, measured like that of the
above-named volcano of Wunzen, in the island of Kiusiu,
by young Japanese observers trained by Siebold, reaches
12,443 feet; almost 300 feet higher, therefore, than the
Peak of Teneriffe, with which Kampfer had already
compared it. (Wilhelm Heine, Eeise nach Japan, 1856,
Bd. ii. S. 4.) The upheaval of this mountain took
place in the fifth year of the reign of Mikado VI.,
(286 B.C.), as described in the following (geologically
remarkable) terms, e In the country about Omi,
a considerable tract of land sank down, a lake was
formed, and the volcano of Fusi rose to view.' The
historically known most violent eruptions were those
of 799, 800, 863, 937, 1032, 1083, and 1707 A.D.,
since which last date the mountain has been in repose.
(2) Volcano Asamajama; the most central of the
active volcanoes in the interior of the country ; distant
80 geographical miles from the S.S.E., and 52 from the
N.N.W. coasts respectively ; in the district Saku, province
Sinano, lat. 36°22/, long. 138° 40' E.; therefore between
the meridians of the two principal towns Mijako and
Jedo. Asamajama had an eruption as early as in the
year 864 (contemporaneously, therefore, with Fusijama).
It had a very violent and destructive eruption in July
1783, since which time it has been in permanent
activity.

" In addition to these volcanoes, European navigators
have observed two small islands with smoking craters,
viz. : — (3) the little island Iwogasima or Iwosima (sima
signifies island, and iwo sulphur ; ga is a mere affix of

358 REACTION OF THE INTERIOR OF THE EARTH

the nominative), He du Volcan of Krusenstern, south of
Kiusiu in Van Diemen Strait, in 30° 43' lat. and 130°
20' E. long., only 54 English miles from the above-
named volcano of Mitake, height 2366 feet. This small
island is mentioned by Linschote in 1596, who says;
' it has a volcano, which is a sulphur or a burning
mountain.' We find it also in the oldest Dutch maps
with the name s Vulcanus.' (Fr. von Siebold, Atlas
vom Jap. Reiche, tab. XI.) Krusenstern saw it smoking
in 1804, as did Captain Blake in 1838, and Gruerin and
de la Roche Poncie in 1846. The height of the cone,
according to the last-mentioned navigator, is 2363 feet.
The small rocky island which Landgrebe mentions as a
volcano, in the Naturgeschichte der Vulkane (Bd. 1, S.
355), according to Kampfer not far from Firator (Fi-
rando), is indisputably Iwosima ; for the group to which
Iwosima belongs is called Kiusiu ku sima, i. e. the nine
islands of Kiusiu, not the 99 islands. There is such a
group near Firato, north of Nagasaki, and nowhere else
in Japan. (4) The island Ohosima (Barnevelt's Islands,
Krusenstern's lie de Vries) ; it is reckoned as belonging
to the province of Idsu in Nippon, and lies in front
of the Bay of Wodawara, in 34° 42' N. lat. and 139° 26'
E. long. Broughton in 1797 saw smoke rise from the
crater, a short time before a violent eruption had taken
place. A range of volcanic islets runs southward from
Ohosima to Fatsi sjo (33° 6' N.), and is continued to the
Bonin Islands (26° 30' N. and 142° 5' E.), which, ac-
cording to Postels (Liitke, Voyage autour du Monde dans
les annees 1826 — 29, t. iii. p. 117), are also volcanic and
subject to very violent earthquakes.

" These then are the eight historically active volcanoes

ON ITS EXTERIOR. VOLCANOES. 359

of Japan proper, in and near the islands Kiusiu and
Nippon. But besides these there is a range of conical
mountains, some of which, characterised by very clearly
marked and often deeply depressed craters, appear to be
long extinct volcanoes : this is the case with the conical
mountain Kaimon, Krusen stern's Pic Homer, in the
most southern part of Kiusiu, on the shore of Van
Diemen Strait, in the province of Satsum (lat. 31° 9'),
scarcely 24 geographical miles S.S.W. from the active
volcano of Mitake ; and so also with Kofusi or the little
Fusi in Sikok, on the small island of Kutsunasima
(province Ijo), lat. 33° 45', on the east shore of the great
strait Suwo Nada or van der Capellen, which separates
the three great portions of the Japanese empire, Kiusiu,
Sikok, and Nippon. On the last, the principal island,
there are enumerated, proceeding from south-west to
north-east, nine such conical mountains, probably trachy-
tic, amongst which the most remarkable is the Sirajama
(White mountain) in the province Kasa (lat. 36° 5'),
which, as well as the Tsjo kaisan, in the province Dewa
(lat. 39° 10'), is estimated to be higher than the lofty
southern volcano Fusijama, more than 12,300 feet high.
Between Sirajama and Tsjo kaisan there is Jakijama
(the flame mountain) in the province of Getsigo, in lat.
36° 53'. The two northernmost conical mountains on
the Tsugar Straits, in sight of the great island of Jezo,
are : (1) Jwakijama, which Krusenstern, whose services
to the knowledge of Japanese geography are deserving of
enduring remembrance, called Pic Tilesius (lat. 40° 42') ;
and (2) Jakejama (burning mountain, lat. 41° 20') in
Nambu, on the north-easternmost end of Nippon, which
has had fiery eruptions since the earliest times."

360 REACTION OF THE INTERIOR OF THE EARTH

On the side of the continent, in the adjacent penin-
sula of Corea or Korai (it is almost united to Kiusiu in
lat. 34° and 34^° by the islands Tsusima and Iki), not-
withstanding its resemblance in form to Kamtschatka,
we as yet know of no volcanoes. The volcanic activity
appears to have been confined to the islands near to it.
Thus, the island-volcano Tsinmura, which the Chinese
call Tanlo, rose from the sea in the year 1007. A
learned man, Tien-kong-tschi, was sent to see and
describe the phenomenon, and to make a drawing of it
(487). On the island Se-he-sure (Quelpaerts of the
Dutch), the mountains show everywhere a volcanic
conical outline. The central mountain, according to La
Perouse and Broughton, reaches a height of 6000 French
feet (6395 feet.) How many more volcanic manifesta-
tions may there not still be to discover in the western
archipelago, where the king of the Corea takes the title
of the king of 10,000 isles !

From Peak Homer (Kaimon-ga-take) at the south-
western point of Kiusiu there runs a bow-shaped
range of small volcanic islands, the opening of the bow
being towards the west, and it comprises between the
straits of Van Diemen and Colnett, Jakunosima and
Tanegasima: and then passing south of Colnett strait
into the Linschote group of Siebold (488) (Archipel Ce-
cille of Captain Guerin, which extends to the parallel
of 29°) it includes Suwasesima, Captain Belcher's Vol-
cano island in 29° 39' N. and 129° 43' E. ; height 2803
feet, according to de la Koche Poncie ; then Basil Hall's
Sulphur island, Torisima or Bird island of the Japanese,
Lung-hoang-schan of Father Graubil, 27° 51' N. and 128°
61' E,, according to the determination of Captain de la

ON ITS EXTERIOR. VOLCANOES. 361

Roche Poncie in 1848. As it is also called Iwosima,
care must be taken not to confound it with the more
northern island of the same name in the Strait of Van
Diemen. It has been extremely well described by
Captain Basil Hall. Between 26° and 27° of latitude
follow the Loo Choo islands, of which Klaproth had
given a special map so early as 1824 ; and more to the
south-west the little archipelago of Madschikosima which
approaches near to the great island of Formosa, and is
regarded by me as the termination of the East Asiatic
islands. In lat. 24° near the east coast of Formosa, in
October 1853, Lieutenant Boyle observed a great vol-
canic eruption in the sea. (Commodore Perry, Exped. to
Japan, vol. i. p. 500.) In the Bonin islands (Buna-sima
of the Japanese, lat. 26|°to 27f°N., 142° 17' E.), Peel's
island has several craters with sulphur and scoriae, which
appear not to have been long extinct. (Perry, vol. i.
p. 200 and 209.)

6. South Asiatic Islands.

We comprise under this head Formosa (Thaywan), the
Philippines, the Sunda Isles, and the Moluccas. The
volcanoes of Formosa were first made known to us by
Klaproth from Chinese sources, which always give such
circumstantial descriptions of natural phsenomena. (489)
They are four in number, one of which, Tschykang, Red
Mountain, has a crater lake of hot water, and has had
great igneous eruptions. The small Baschi islands, and
the Babuyans which so late as 1831, according to Mayen,
had a vehement fiery eruption, connect Formosa with
the Philippines, the smallest and most broken of which

362 REACTION OF THE INTERIOR OF THE EARTH

are the principal volcanic sites. Leopold von Buch enu-
merates in the Philippines nineteen lofty isolated conical
mountains, all called in the country "Volcanes," but
of which some probably are closed trachytic domes.
Dana thinks that in the southern part of Luzon there
are now only two active volcanoes, Taal in the Laguna
de Bongbong or Bombon, with an encircling ridge
enclosing a second Laguna (see above, p. 242.) ; and in
the south part of the peninsula of Camarines the vol-
cano of Albay or Mayon, which the natives call Isaroe ;
it is about 3200 feet high, and had great eruptions in
the years 1800 and 1814. In the northern part of
Luzon granite and mica slate, and even sedimentary
formations and coal, are diffused. (49°)

The long extended group of the Sulu or Solo islands,
probably fully a hundred in number, connecting Minda-
nao with Borneo, is partly volcanic and partly intersected
by coral reefs. Isolated unopened trachytic conical
peaks are, indeed, often called "Volcanes" by the
Spaniards.

Dr. Junghuhn, after having carefully passed in re-
view all those islands which are to the south of the fifth
parallel of north latitude (south, therefore, of the
Philippines), and between the meridians of the Nico-
bars and the north-west of New Oruinea, states as the
result that "in a wreath of islands surrounding the
great and almost continental island of Borneo there
are 109 lofty fire-emitting mountains and ten mud-
volcanoes." This is by no approximate estimate but by
actual enumeration.

In Borneo itself, the Griava Maggiore of Marco
Polo (491), we have as yet no certain knowledge of any

ON ITS EXTERIOR. VOLCANOES. 363

active volcano; we are indeed only acquainted with
some narrow strips of coast (on the north side as far as
the little island of Labuan and Cape Balambangan, on
the west coast to the mouth of the Pontianak, and on
the south-easternmost point the district of Banjermas-
Sing on account of its gold, diamond, and platina
washings) : nor is the loftiest mountain of the whole
island, (perhaps the loftiest in any of the South
Asiatic islands,) Kina Bailu, with its two summits, of
which the northern is only about thirty-two geogra-
phical miles from the Pirates' Coast, supposed to be a
volcano; Captain Belcher found it about 13,700 feet
high, or nearly 4000 feet higher than Grunung Pasaman
(Mount Ophir), in Sumatra. (492) But on the other
hand, Rajah Brooke mentions a much lower mountain
in the Province of Sarawak as bearing the name of
Grunung Api (Fire Mountain in Malay), which name
and the scoriae which surround the mountain would
indicate a former state of volcanic activity. Great
deposits of auriferous sand between quartzose dykes,
much tin washed down on both banks of the river, and
the felspathic porphyry (493) of the Serambo mountains,
indicate a great diffusion of so-called primitive and
transition rocks. From the only well-assured determi-
nations which we possess from a well trained geologist,
(Dr. Ludwig Homer, son of the meritorious Zurich
astronomer and circumnavigator,) washings are carried
on at several places in the south-east of Borneo, in
which process gold, diamonds, platina, osmium, and
iridium are found, as in the Siberian Ural (but hitherto
no palladium). A range of hills 3410 feet high, the

364 REACTION OF THE INTERIOR OF THE EARTH

Ratuh mountains, is very near, and has formations of
serpentine, gabbro, and syenite. (494)

In the other three great Sunda Isles, Sumatra, Java,
and Celebes, Junghuhn counts of active volcanoes; — in
Sumatra 6 or 7, in Java 20 or 23, and in Celebes 11;
to Flores he assigns 6. We have already spoken in
detail, p. 280 to 288, of the volcanoes of Java. In
Sumatra, which has not yet been thoroughly explored,
out of nineteen conical mountains which have a vol-
canic appearance, six are active volcanoes. (495) Those
recognised as such are : Gunung Indrapura, about 11,500
French feet (12,256 English) according to angles of
altitude taken at sea (perhaps its height is similar to
that of the more exactly measured G. Semeru or Maha-
Meru in Java); Grunung Pasaman, also called Ophir,
which was ascended by Dr. L. Horner (9602 feet), with
an almost extinct crater.; Grunung Salasi rich in sulphur;
it erupted scoriae in 1833 and 1845 ; Gunung Merapi,
(9570 feet) also ascended by Dr. L. Horner in company
with Dr. Korthal in 1834, the most active of all the
volcanoes of Sumatra, and not to be confounded with
the two mountains of the same name in Java (496) ;
Gunung Ipu, a truncated cone which sends forth smoke;
and Gunung Dempo, in the inland district of Bencoolen;
its height is estimated at 10,000 French feet (about
10,660 English).

Four small islands, cones of trachyte, of which Peak
Rekata and Panahitam (the Prince's islands) are the
highest, rise in the Strait of Sunda and connect the
Sumatra range of volcanoes with that of Java ; and in a
similar manner, at the eastern end of Java, its eastern-
most volcano, Idgien, is connected by the active volcanoes

ON ITS EXTERIOR. VOLCANOES. 365

Grunung Batur and Grunung Ajung on the adjacent
island of Bali with the long chain of the Lesser Sunda
Isles. We have in succession, proceeding eastward from
Bali, in the island of Lombok, the smoking volcano of
Rindjani, according to Mr. Melville de Carnbee's trigo-
nometrical measurement 12,363 feet; Temboro (5862
feet), in Sumbawa or Sambawa island, whose eruption of
pumice and ashes which darkened the air, in April 1815,
is one of the greatest of which history has preserved
the record (497) ; and six partly still smoking conical
volcanoes in the island of Flores.

The great and many-armed island of Celebes contains
six volcanoes which are not yet all extinct. They lie
close together in the north-easternmost narrow peninsula
of Menado. Sulphurous springs gush forth boiling hot
in their vicinity. My Piedmontese friend, Count Carlo
Vidua, a very extensive and observant traveller, unhap-
pily fell into one of these springs situated near the
route from Bonder to Lamovang, and died from the
effects of the inflamed wounds caused by the scalding
mud. Like the little island of Banda, one of the
Moluccas, which consists of the volcano Grunung Api
(only about 1800 feet high, and which was active from
1586 to 1824), the larger island of Ternate also consists
solely of the cone of Chining Q-ama Lama (5755 feet
high), whose violent eruptions from 1838 to 1849 are
described as having taken place (after an interval of
more than a century and a half of entire repose) at ten
distinct epochs. Junghuhn tells us that, in the eruption
of the 3rd of February 1840, a stream of lava issued
from a fissure near Fort Toluko and flowed down to the
sea-shore (49C<); he is uncertain "whether the lava may

366 REACTION OF THE INTERIOR OF THE EARTH

have formed a connected completely molten mass, or
whether it may have issued forth in the form of glowing
fragments, which rolled down, and as they followed in
close succession pressed each other over the more level
ground." If to the more important volcanic cones
which have been named above, we add the many very
small island-volcanoes which cannot be here enumerated
in detail, then the sum total of all the volcanoes situ-
ated to the south of the parallel of Cape Serangani, on
Mindanao, one of the Philippines, and between the meri-
dians of the north-west cape of New Guinea on the east,
and the group of the Nicobar and Andaman islands on
the west, amounts, as has been already said, to 109.(499)
It is not, however, meant that these are now all active
volcanoes. The estimation has been made in such
manner that "in Java it includes 45 volcanoes, most of
them being conical and furnished with craters." But of
these 45 only 21, and of the whole number of 109 only
from 42 to 45 are now active, or are known to have
been so within historic times. The great peak of Timor
once served, like Stromboli, as a lighthouse to mariners.
On the small island Pulu Batu (also called P. Komba),
a little to the north of Flores, a volcano was seen in
1850 to pour down glowing lava on the sea-beach, as
was also the case with the Peak on the larger Sangir
island between Magindanao and Celebes in 1812, and
again quite recently in the spring of 1856. Junghuhn
doubts whether the conical Mount Wawani or Ateti in
Amboina poured forth anything more than hot mud in
1674, and he regards the island as at present belonging
only to the class of solfktaras. The great group of the
South Asiatic islands connects itself on the one hand by

ON ITS EXTEEIOR. VOLCANOES. 367

the section of the western Sunda Isles with the Nicobar
and Andaman islands of the Indian Ocean, and on the
other hand by the section of the Moluccas and Philip-
pines with the Papuas, Pelew islands, and Carolinas of
the Pacific. But we will consider next the less nume-
rous and more scattered groups of the Indian Ocean.

7. The Indian Ocean.

The Indian Ocean comprises the space between the
west coast of the peninsula of Malacca and the east
coast of Africa, including therefore the Bay of Bengal
and the Arabian and Ethiopian seas. We will follow
the direction of the volcanic activity in the Indian
Ocean, which is from north-east to south-west.

Barren island, in the Bay of Bengal, a little to the
east of the greater Andaman island (12° 15' N.), is
rightly termed an active cone of eruption rising out of a
crater of elevation. The sea flows in through a narrow
aperture and fills an interior basin. The appearance of
this island, which was discovered by Horsburgh in 1791,
is exceedingly instructive for the theory of the formation
of volcanic frameworks. We see here completed and
permanent that which, at Santorin and at other points
of the globe, Nature had presented to our observation
only transitorily. (50?) The eruptions in November 1803
were, like those of Sangay in the Cordilleras of Quito,
very decidedly periodical, with intervals of ten minutes.
(Leop. von Buch, in the Abhandl. der Berlin Akademie,
1818-1819, S. 62).

The island of Narcondam (13° 24' N.), north of Bar-
ren island, has also shown volcanic activity in former

368 REACTION OF THE INTERIOR OF THE EARTH

times, as has also the still more northerly conical moun-
tain of the Island of Cheduba (18° 52' N.), near the
coast of Arracan (Silliman's American Journal, vol. 38,
p. 385).

The most active volcano, if the abundance of lava
poured forth be made the criterion, not only in the
Indian Ocean but almost in the whole of that part of
the southern hemisphere which lies between the meri-
dians of the west coast of New Holland and the east
coast of America, is the volcano of the Island of Bour-
bon in the group of the Mascareignes. The greater
part of the island, in particular the western and inland
portion, is basaltic. More recent basaltic dykes, poor in
olivine, intersect the older rock which is rich in olivine ;
and beds of lignite are enclosed in basalt. The culmi-
nating points of the mountainous island are Le Grros
Morne and Les Trois Salazes whose height was over-
estimated by Lacaille at 10,000 French feet (10,658
English). The volcanic activity is now confined to the
south-eastern part of the island, Le Grand Pays Brule.
The summit of the Volcano of Bourbon, which almost
every year sends forth, according to Hubert, two streams
of lava, which often reach the sea, is, according to
Berth's measurement, 8000 feet high.{501) It shows
many cones of eruption to which distinct names have
been given, and of which sometimes one and sometimes
another sends forth eruptions. The summit itself does
so but rarely. The lavas contain glassy felspar, and are
therefore trachytic rather than basaltic. The showers
of ashes often contain olivine in the form of long and
fine threads, a phenomenon which is also found in the
Volcano of Owyhee. A great eruption which took

ON ITS EXTERIOR. VOLCANOES. 369

place in 1821 covered the whole Island of Bourbon
with these glassy threads.

In the great Terra Incognita of Madagascar, we know
only of the wide distribution of pumice at Tintingue,
opposite to the French He Sainte Marie, and basalt
showing itself south of the Bay of Diego Suarez, near
the northernmost Cap d'Ambre, surrounded by gneiss
and granite. The height of the southern central range
of the Ambohistmen mountains is estimated, probably
with great uncertainty, at between ten and eleven thou-
sand feet. To the west of Madagascar, at the northern
outlet of the Mozambique channel, the largest of the
Comoro islands has a burning volcano (Darwin, Coral
Eeefs, p. 122).

The small volcanic island of St. Paul (in 38° 38' S.),
south of Amsterdam, is called volcanic not only on
account of its shape which vividly recalls that of San-
torin, Barren island, and Deception island in the New
Shetland group, but also on account of the repeatedly
observed eruptions of fire and vapour in recent times.
The very characteristic drawing given by Valentyn, in
his work on the Banda Isles, on the occasion of the
expedition of Willem de Vlaming (Nov. 1696), agrees
perfectly, as does also the assigned latitude, with the
drawings in the Atlas of Lord Macartney's expedition,
and with Captain Blackwood's survey in 1842. The
crater-shaped circular bay, almost an English mile
broad, is every where surrounded (with the exception of
one narrow opening through which the sea flows in at
high tide) by a high rocky escarpment, having a verti-
cally precipitous face internally, and a gentle and
gradual declivity externally. (502)
VOL. iv. B B

370 REACTION OF THE INTERIOR OF THE EARTH

The Island of Amsterdam (37° 48' S.), or 50' of lati-
tude to the north of St. Paul's, consists, according to
Valentyn's drawing, of a single, wooded, rather rounded
mountain, on the highest part of which rises a small
cubical rock almost similar to that on the Cofre de
Perote in Mexico. In the expedition of D'Entrecasteaux,
in March 1792, the island was seen for two days wrapped
in fire and smoke. The smell of the smoke seemed to
indicate a forest and earth fire ; the voyagers did indeed
think they saw columns of smoke rising here and there
out of the ground near the shore ; but the naturalists
who accompanied the expedition came to the conclusion
that the phenomenon was at least not to be ascribed to
the eruption (503) of the high mountain as a volcano,
We might adduce as more secure evidences of a more
ancient and genuine volcanic activity in the Island of
Amsterdam, the beds of pumice (uitgebranden puim-
steen) mentioned already by Valentyn, according to
Vlaming's ship's journal in 1696,

To the south-east of the extremity of Africa there
are Marion's or Prince Edward's island (47° 2' S.) and
Possession island, belonging to the Crozet group (46° 28'
S. lat., and 51° 58' E. long.). Both show traces of
former volcanic activity ; they are small conical hills (504)
with eruption-orifices surrounded by columnar basalt.

To the eastward, in almost the same latitude, follows
Kerguelen island (Cook's Island of Desolation), of which
we owe the first geological description to Sir James
Koss's antarctic expedition, so fruitful in successful
results. At the place called by Cook Christmas Har-
bour (48° 41' S. lat., 69° 04' E. long.), basaltic lavas se-
veral feet thick envelope fossil trunks of trees. The

ON ITS EXTERIOE. VOLCANOES. 371

picturesque "Arched Kock" is a natural passage through
a narrow projection of basaltic rock. There are in the
vicinity conical mounts, of which the highest rise to
2664 feet, with extinct craters; masses of greenstone
and porphyry traversed by dykes of basalt, and amyg-
daloid with quartzose drusic cavities, at Cumberland
Bay. Most remarkable are the many beds of coal em-
bedded in trap-rock (dolerite as at Meissner in Hesse ? ),
varying in thickness from a few inches to four feet. (505)
If we cast a general glance over the domain of the
Indian Ocean, we see that the extremity of the Sunda
range, which in Sumatra has assumed a north-westerly
curvature, is prolonged through the Nicobars, the
greater and lesser Andamans, and the volcanoes of
Barren island, Narcondam and Cheduba, into the eastern
part of the Bay of Bengal, in a line almost parallel to
the coast of Malacca and Tenasserim. The western part
of the Bay of Bengal, along the coasts of Orissa and Co-
romandel, is without islands ; Ceylon, like Madagascar,
has rather a continental character. On the western side
of the great Indian peninsula, over against the range of
the Neilgherries and the coasts of Canara and Malabar,
the three groups of the Laccadives, Maldives, and Cha-
gos, form in a north and south direction a chain of
islands from 14° N. to 8° S. latitude connecting itself
through the banks of Sahia de Malha and Cargados
Carajos with the volcanic group of the Mascareignes and
with Madagascar ; they are all, so far as is visible, struc-
tures raised by the coral polypes, true atolls or lagoon
reefs ; according to Darwin's ingenious conjecture, that
we have here a wide space of sea bottom forming an
area not of elevation but of subsidence.

BB 2

REACTION OF THE INTERIOR OF THE EARTH

8. The Pacific.

>mpare the part of the earth's surface which
is at present covered by water with the area of dry land
(the ratio being approximately (506) as 2 '7 : 1), we may
be struck by surprise in geological respects at the rarity
of still active volcanoes in the oceanic region. The

O

Pacific Ocean, whose surface is nearly one sixth greater
than that of the whole dry land of our planet, — whose
breadth in the equatorial region, from the Gralapagos to
the Pelew islands, is nearly two fifths of the whole cir-
cumference of the globe, — presents fewer smoking vol-
canoes, fewer openings through which the interior of
the planet still maintains an active communication with
its atmospheric envelope, than does the single island of
Java. The geologist of the great American exploring
expedition under the command of Charles Wilkes
(1838-1842), the ingenious James Dana, has un-
doubtedly the great merit of having shed a new light
on the whole Pacific world of islands, by taking as a
foundation his own researches combined with a careful
assemblage and comparison of all well-assured previous
observations, and proceeding thence to a generalisation
of views repecting the shape, distribution, and direction
of axis in all the various groups of islands ; the charac-
ter of the rock ; and periods of subsidence and elevation
of extensive tracts of the bottom of the sea. If I avail
myself largely of his writings as well as of the admir-
able investigations of Charles Darwin, the geologist of
Captain Fitz-Koy's expedition in 1832-1836, without
special reference on each occasion, the high esteem
which I have for so many years expressed for both those

ON ITS EXTERIOR. YOLCANOES. 373

able men will sufficiently obviate the possibility of any
misunderstanding.

I prefer to avoid the sectional names arbitrarily and
variously given upon grounds respecting either the
number and size of islands, or the complexion and de-
scent of their inhabitants, Polynesie, Micronesie, Mela-
nesie, and Malaisie (507) ; and begin the enumeration of
still active volcanoes in the Pacific or South Sea wi^h
those which are to the north of the equator ; and will
then proceed to consider, in a direction from east to
west, the islands lying between the equator and the
parallel of 30° S. The many basaltic and trachytic
islets with their countless craters, eruptive formerly but
at very different periods, must not indeed be regarded
as dispersed without any regularity or definite order. (508)
We can recognise in the greater number that their ele-
vation has taken place on extended fissures and subma-
rine mountain-ranges, following determinate directions
in regions and groups belonging to different systems, as
we have seen to be the case in the continental mountain-
ranges of Central Asia and the Caucasus ; but the rela-
tive positions of the very limited number of volcanic
apertures which show themselves contemporaneously
active at any particular epoch, are probably dependent
on very local disturbances affecting the conducting
fissures. If we attempt to draw lines through three
volcanoes which are all active at the present time and
are 2400 and 3000 geographical miles from each other,
without any intermediate cases of eruption, — (I mean
the volcanoes of Mauna Loa, with Kilauea on its eastern
declivity, the conical Mount Tanna in 'the New Hebrides,
and Assumption in the northern Ladrones,) — such lines

374 REACTION OF THE INTERIOR OF THE EARTH

could give us no information as to the general genetic
connection of the volcanoes in the basin of the Pacific.
It is otherwise when we confine our consideration to
single groups of islands, and go back in imagination to
the, perhaps, pre-historic epochs, when the many now
extinct linearly arranged craters of the Ladrones (Ma-
rianas), the New Hebrides, and Solomon's islands
were in a state of activity; but which craters we are
certainly not justified in assuming to have become suc-
cessively extinct one after another in any particular
geographical direction, as, for example, from south-east
to north-west, or from north to south. I have here
spoken of volcanic ranges of islands in the high seas,
but the same reasoning would apply to the Aleutian or
other "coast islands." Creneral conclusions respecting
the direction of a process of cooling are illusory, on
account of the influence on such cooling processes of
temporary obstruction or freedom from obstruction in
the channels of conduction.

Mauna Loa, or Boa (or Mouna acording to the Eng-
lish mode of writing), which was found by a careful
measurement (509) of the American exploring expedition
under Captain Wilkes to be 13,758 feet high, or 1600
feet higher than the Peak of Teneriffe, is both the
greatest volcano in the Pacific, and also is the only one
still thoroughly active in the entire volcanic archipelago
of the Hawaii or Sandwich islands. The summit-craters,
of which the largest is about 13,000 feet in diameter,
present in their ordinary .state a solid floor of cooled
lava and scoriae from which rise small cones of eruption
exhaling vapours. The summit-openings are in general
but little active; but in June 1832, and in January 1843,

ON ITS EXTERIOE. VOLCANOES. 375

eruptions took place from them which lasted many
weeks, and even sent forth streams of lava from 20 to
28 geographical miles in length, reaching to the foot of
Mauna Kea. The angle of inclination of the connected
flowing stream (51°) was for the most part 6°, often
from 10° to 15°, and even 25°. A very remarkable fea-
ture in Mouna Loa is the absence of any cone of ashes
or cinders like the Peak of Teneriffe, Cotopaxi, and so
many other volcanoes; pumice is also almost entirely
wanting (511); although the blackish-grey, rather
trachytic than basaltic, lavas of the summit are rich in
felspar. The extraordinary fluidity of the lavas of
Mouna Loa, whether from the summit-crater (Mokua-
weo-weo), or from the lake of lava (situated at a height
of only 3969 feet above the sea on the eastern declivity
of the volcano), is evidenced by the sometimes smooth
and sometimes curled threads of glass which the wind
carries and scatters over all the island. This hair-like
glass, which is also emitted from the volcano of Bourbon,
is called in Hawaii (Owyhee), " Pele's Hair," from the
protecting goddess of the country.

Dana, with great sagacity, has shown that Mouna
Loa is not a central volcano to the Sandwich island
group, and that the lava-lake Kirauea or Kilauea is not
a solfatara. (512) The longer diameter of the basin of
Kilauea is 16,000 feet, and its shorter diameter 7460
feet. The steaming, boiling, heaving fluid of this lava-
pool, does not under ordinary circumstances occupy the
whole cavity, but only a space a little less than 14,000
feet long and rather more than 5000 feet broad. There
is a descent by a kind of steps down the crater margin.
This grand phenomenon leaves on the mind a wonderful

376 REACTION OF THE INTERIOR OF THE EARTH

impression of stillness and solemn repose. The approach
of an eruption is not here announced by earthquakes or
subterranean noises, but solely by a sudden rising and
falling of the surface of the lava, sometimes from a
depth of 300 or 400 feet to the upper margin of the
basin, and the reverse. If, disregarding the enormous
difference of size, we should be disposed to compare the
colossal basin of Kilauea with the small lateral craters
(first circumstantially described by Spallanzani) situated
on the declivity of Stromboli, at four fifths of its height,
and of which the larger are only about 200 and the
smaller about 30 feet across, we should remember the
important distinction constituted by the fact, that these
fiery openings of Stromboli (which is itself unopened
at the summit) throw up scoriae to a great height, and
even pour forth lavas. Although the great lava-lake of
Kilauea (which may be termed the lower and secondary
crater of the active volcano of Mouna Loa) may some-
times threaten to overflow its margin, yet it never has
actually so overflowed as to give birth to a proper stream
of lava. Such streams are, however, formed by the
descent of the lava through subterranean channels, and
by the formation of new openings of eruption at a
distance of 16 or 20 miles from the fiery lake at points
much lower down. After such eruptions, occasioned
by the pressure of the enormous mass of lava in
Kilauea, have taken place, the surface of the lava in
the basin sinks to a lower level. (513)

Of the other two high mountains in Hawaii, Mouna
Kea and Mouna Hualalai, the first is, according to
Captain Wilkes, 192 feet higher than Mouna Loa: it is
a mountain cone whose summit no longer presents

ON ITS EXTERIOR. VOLCANOES. 377

a terminal crater, but only hills of long extinguished
scoriae. Mouna Hualalai* is about 10,000 feet high, and
is still burning. In 1801 an eruption took place in
which the lava reached the sea on the western side.
It is to the elevation from the bottom of the sea of
the three great mountains of Loa, Kea, and Hualalai,
that the whole of the island of Hawaii owes its origin.
In the descriptions of the many ascents of Mouna Loa,
among which that given by Captain Wilkes' Expedition
rests on examinations continued during 28 days, a
fall of snow with a temperature of from 9° to 14°
below the freezing point is spoken of, and single patches
of snow near the summit are said to have been seen
from a distance by the aid of a telescope ; but perpetual
snow is never mentioned. (5l4) I have already remarked
that according to the measurements which we may now
regard as most accurate, 13,759 feet for Mouna Loa, and
13,951 feet for Mouna Kea, those mountains are fully
1000 and 800 feet below what I found to be the lower
limits of perpetual snow on the continental mountains of
Mexico in 19^° of latitude. We should expect to find
the limit of perpetual snow rather lower down on a small
island than on a continent, on account of the influence of
the surrounding ocean in occasioning, in the hot season,
a lower temperature of the inferior atmospheric strata,
and the presence of a greater quantity of aqueous
vapour in the higher portions of the atmosphere over
the island.

The volcanoes of Tafoa * and Amargura *, in the
Tonga group, are both active, and the latter sent forth
a considerable flow of lava on the 9th of July 1847. (515)
It is an exceedingly remarkable circumstance, and in

378 REACTION OF THE INTERIOR OF THE EARTH

complete accordance with the experience that the coral
animals avoid the coasts of volcanoes which now are or
have recently been in a state of activity, that the Tonga
islands (rich in coral-reefs) of Tafoa and of the cone of
Kao are quite stripped of living coral. (516)

The volcanoes of Tanna * and Ambrym * come next ;
the latter is to the west of Mallicollo, in the group of the
New Hebrides. The volcano of Tanna, which was first
described by Reinhold Forster, was found in full erup-
tion on Cook's first discovery of the island in 1774. It
has continued active ever since, and as its height is
scarcely 460 feet, it must be ranked, together with
the Japanese Kosima and the volcano of Mendana
mentioned below, as belonging to the class of the least
elevated of fire-emitting mountains. On Mallicollo is
much pumice.

Mathews Rock* is a small island-rock about 1180
feet high, which sends forth smoke, and which D'Urville
observed in eruption in January 1828. It is to the
east of the south point of New Caledonia.

The volcano of Tinakoro *, in the Vanikoro or Santa
Cruz group.

In the same archipelago of Santa Cruz, fully 80 geo-
graphical miles N.N.W. of Tinakoro, there rises from
the sea a small volcano * only about 200 feet in height
(under 200 French or 213 English feet), which was first
seen by Mendana in 1595. It is in 10° 23' S. Its fiery
eruptions have sometimes been periodical, occurring
every ten minutes; and sometimes, as when seen by
D'Entrecasteaux's expedition, the column of vapour
may be said to have been itself the crater.

In the Solomon group there is an active volcano on

ON ITS EXTERIOR. VOLCANOES. 379

the island Sesarga.* Near to it, and therefore near also
to the south-eastern extremity of the long range
of islands, and in the direction of the Vanikoro or
Santa Cruz group, outbreaks of volcanic activity have
been remarked on the coast of Gruadalcanar.

In the Ladrones, or Marianas, in the northern part of
that range of islands which seems to have been elevated
from a fissure running in the direction of the meridian,
Gruguan*, Pagon *, and the Volcan Grande of Asuncion *
are said to be still active.

The direction of the coast-line of the small continent
of New Holland, and especially the alteration of direc-
tion undergone by its east coast in 25° S. (between Cape
Hervey and Moreton Bay) appears reflected, or repeated,
in the zone of neighbouring eastern islands. The great
southern island of the two islands of New Zealand, and
the Kermadec and Tonga groups have a general south-
west and north-east direction : while on the other hand
the northern part of the northern island of New Zealand,
from the Bay of Plenty to Cape Oton, New Caledonia
and New Guinea, the New Hebrides, the Solomon
Islands, New Ireland and New Britain, all follow a
south-east and north-west direction, for the most part
N. 48° W. Leopold von Buch (517) was the first to
call attention to these relations between continental
masses and the neighbouring islands, in the case
of the archipelago of Greece, and of the coral-sea of
Australia. In the islands of this last-named sea also, as
was already remarked both by Cook's companion, Forster,
and La Billardiere, granite and mica-slate, the quartzose
rocks which were once termed primitive, are not wanting.
Dana has also found them in the northern island of

B80 REACTION OF THE INTERIOR OF THE EARTH

New Zealand, to the west of Tipuna, in the Bay of
Islands. (518)

At the southern extremity of New Holland (Australia
Felix), at the foot and to the south of the Grampians,
fresh traces of ancient volcanic action are found : and to
the north-west of Port Phillip there are, according
to Dana, a umber of volcanic cones and lava beds,
as also towards the Murray Eiver (Dana, p. 453).

In New Britain*, near its eastern and its western coasts,
at least three cones have been observed as burning and
yielding lavas within historic times, by Tasman, Dampier,
Carteret, and La Billardiere.

There are two active volcanoes on the north-east coast
of New Guinea, opposite to New Britain and the Admi-
ralty islands, which are rich in obsidian.

In New Zealand, where the geology of the northern
island at least has been well elucidated by the important
work of Ernst Dieffenbach and the fine investigations of
Dana, basaltic and trachytic rock breaks through the
generally diffused plutonic and sedimentary rocks at
several points; for instance, in an exceedingly small
area near the Bay of Islands in 35° 2' S., where rise the
cones of ashes, crowned by extinct craters, of Turoto
and Poerua; so also, further to the south (between 37J°
and 39^° S.), a strip of volcanic ground extends across
the whole of the northern island, a distance of more
than 160 miles from north-east to south-west, from the
eastern Bay of Plenty to the western Cape Egmont.
Here, as we have seen on a larger scale on the American
continent (in Mexico), this zone of volcanic activity
appears as a cross fissure running from sea to sea, and
intersecting at a considerable angle the long inland

ON ITS EXTERIOR. VOLCANOES. 381

chain of mountains which seems to give the form to the
entire island. On the volcanic cross fissure, and as it
may seem at the point of intersection, we have the
high cone of Tongariro* (6198 feet), whose crater on
the cone of ashes has been reached by Bidwell; and,
rather more to the south, Kuapahu (9005 feet). The
north-east extremity of the zone is formed by a con-
stantly smoking solfatara, the island-volcano of Puhia-i-
wakati* (519) (White island, in the Bay of Plenty, lat.
38i.° S.). Keturning to the south-west, we have first,
on the coast itself, the extinct volcano of Putawaki
(Mount Edgecombe), 9630 feet high; a snow-clad moun-
tain, and probably the loftiest in New Zealand; and
inland, between Mount Edgecombe and the still burning
Tongariro*, which has poured forth some streams of
lava, we find a long chain of lakes of which some are of
boiling water. The lake of Taupo, which is surrounded
by fine shining leucite and sanidine sand, and by hills
of pumice, is nearly twenty-four miles in length ; it is
in the middle of the northern island of New Zealand,
and is 1337 feet above the level of the sea, according to
Dieffenbach. The space around, for two English square
miles, is covered with solfataras, caves emitting vapours,
and thermal springs ; which last, like the geysirs in Ice-
land, form a variety of deposits of silicates. (52°) To
the west of Tongariro*, the principal seat of volcanic
activity, whose crater still emits vapours and pumice
ashes, and which is only sixteen geographical miles
distant from the western sea, rises the volcano of Tara-
naki (Mount Egmont), 8838 feet high and which was
first ascended and measured by Dr. Ernst Dieffenbach
in November 1840. The summit of the cone, of which

382 KEACTION OF THE INTERIOR OF THE EARTH

the outline resembles that of Tolima more than that of
Cotopaxi, terminates in a plateau from which there
rises a very steep cone of ashes. No traces of present
activity, as on the volcano of White Island* and that of
Tongariro*, have been observed ; nor any connected
lava-streams. Masses of a ringing kind of rock lami-
nated in very thin scales, and which project, as on one side
of the Peak of Teneriffe, in sharp ridges from among
scorise of the cone of ashes itself, resemble porphyritic
schist (or phonolite).

A narrow, long-extended, uninterrupted series of
island groups over south-east and north-west fissures,
as New Caledonia and New Ghiinea, the New Hebrides
and Solomon Islands, Pitcairn, Tahiti, and the Paumotu
islands, cross the great Pacific Ocean in the southern
hemisphere between the parallels of 12° and 27° S.,
extending in an east and west direction over 5400 geo-
graphical miles, from the meridian of the east coast of
Australia to Easter Island and the Rock Sala y Gomez.
The western portions of this great congeries of islands
(New Britain*, the New Hebrides*, Vanikoro* in the
Santa Cruz archipelago, and the Tonga* group) show
at the present time (the middle of the 19th century)
igneous activity. New Caledonia, although surrounded
by basaltic -and other volcanic islands, has merely pin-
tonic rocks (521), as is the case in the Azores with Santa
Maria (522) according to Leopold von Buch, and also
with Flores and Graeiosa according to Count Bedemar.
To this absence of volcanic activity in New Caledonia,
where sedimentary formations and beds of coal have
lately been discovered, is ascribed the great development
of live coral-reefs. The group of the Viti or Feejee

ON ITS EXTERIOR. VOLCANOES. 383

islands is both basaltic and trachytic, yet in the point of
view now under consideration it is only distinguished by
hot-springs in Savu Bay in the island of Vanua Leboo.(523)
The Samoa group (Navigator's islands), north-east of
the Feejee, and almost due north of the still volcanically
active Tonga group, is also basaltic, and is characterised
by a great number of linearly arranged craters of erup-
tion surrounded by beds of tufa with inbaked pieces of
coral. Most remarkable, geologically, is the Peak of
Tafua in the island of Upolu belonging to the Samoa
group, and not to be confounded with the still burning
Peak of Tafoa south of Amargura in the Tonga group.
The Peak of Tafua, 2138 feet high, which was first
ascended and measured by Dana (524), has a large crater
quite filled with forest trees and crowned by a regularly
rounded cone of ashes. There are no traces of lava-
streams, but there are scoriaceous lava-fields (Malpais of
the Spaniards), with curled and often reticulated surface
of the lava, at the conical Mountain of Apia (2576 feet),
also in the island of Upolu, as well as at the Peak of
Fao which reaches a height of 3000 French (3197 Eng.)
feet. The lava-fields of Apia contain narrow subter-
rannean caves.

Tahiti in the middle of the Society islands is much more
trachytic than basaltic; it shows, properly speaking,
only the ruins of its former volcanic framework ; and
amidst these grand remains, forming sometimes lofty
walls and sometimes jagged peaks, with vertical preci-
pices several thousand feet in depth, it is difficult to
decipher the original form of the volcano. Of the two
highest summits, Aorai and Orohena, Aorai was ascended
first by Dana (525), and examined by that sound geolo-

384 REACTION OF THE INTERIOR OF THE EARTH

gist. The trachytic mountain Orohena is said to be as
high as Etna. Tahiti has, therefore, next to the active
group of the Sandwich islands, the loftiest eruptive
rock in the whole of the great oceanic domain lying
between1 the continents of America and Asia. A fel-
spathic rock of the small islands of Borabora and Mau-
rua in the vicinity of Tahiti, which later travellers
have termed syenite, and which Ellis in his " Polynesian
Kesearches" designates as a granitic aggregate of felspar
and quartz, is deserving of a much more exact minera-
logical examination ; as porous scoriaceous basalt breaks
forth in near proximity to it. Neither extinct craters
nor streams of lava are now found in the Society islands.
It is natural to ask, have the craters on the mountain-
summits been destroyed? or are the now riven and
altered remains of the volcanic frameworks the ruins of
closed domes ? or may we suppose that here, as has pro-
bably been the case at many other points of the up-
heaved sea bottom, basalt and trachyte beds have been
poured forth directly from earth-fissures? Extremes of
great tenacity or of great fluidity in the substances
which issued forth, as well as differences of narrowness
or breadth in the fissures through which they have
issued, modify the shape assumed by the volcanic rocks
formed by them; and where friction produces trituration
into small fragments and into what are then called
ashes, small, and most often transitory, cones of erup-
tion arise, which are not to be confounded with the
great terminal cones of ashes of permanent frameworks.
At only a short distance to the east of the Society
islands there follow the Low islands, or Paumotu.
These are mere coral-islands, with the remarkable ex-

ON ITS EXTERIOR. VOLCANOES. 385

ception of the small basaltic group of Gambler and
Pitcairn islands. (52G) Volcanic rock, as in these last,
is also found in the same parallel (between 25° and 27°
S.) 1260 geographical miles more to the eastward in
Easter Island (Waihu), and probably also 240 miles
further, in the Sala y Gomez rock. On Waihu, where
the highest conical summit is only about 1000 feet
high, Captain Beechey observed a range of craters, none
of which, however, appeared to be burning.

The islands of the Pacific at their extreme eastern
termination on the side of the American continent con-
clude with the group of the Galapagos; one of those
in which volcanic activity is most developed. Scarcely
anywhere in so small a space, barely 120 or 140 geogra-
phical miles in diameter, have so many conical mountains
and extinct craters (traces of ancient communication be-
tween the interior and the atmosphere) remained visible.
Darwin estimates the number of craters at almost
2000. When he visited these islands in the expedition
of the Beagle under Captain Fitz-Roy, two craters were
simultaneously in fiery eruption. In all the islands,
streams of very fluid lava may be seen separating into
branches, and often reaching the sea. Almost all are
rich in augite and olivine, and some which are more
trachytic are said to contain albite (527) in large crystals.
It would be well in the present state of improved mine-
ralogical knowledge to examine whether these porphy-
ritic trachytes do not contain oligoclase, as at Teneriffe
and in Popocatepetl and Chimborazo ; or labradorite, as
in Etna and Stromboli. Pumice is entirely wanting in
the Galapagos, as in Vesuvius, as a production of the
volcano itself ; neither is hornblende ever mentioned as

VOL. iv. C c

386 REACTION OF THE INTERIOR OF THE EARTH

having been found in them, so that we have not here
the trachyte formation of Toluca, Orizaba, and of some
of the volcanoes of Java, from which Dr. Junghuhn
sent me selected solid pieces of lava to be examined
by Gustav Eose. In Albemarle island, which is the
largest and westernmost of the Galapagos, the conical
mounts are arranged linearly, and we may infer there-
fore over fissures. Their greatest height, however, does
not exceed 4636 feet. The western bay, in which rises
the island peak of Narborough which in 1825 was in
vehement eruption, is described by Leopold von Buch (528)
as a crater of elevation, and is compared to Santorin.
Many of the margins of craters in the Galapagos are
formed of beds of tufa which fall away on all sides. It is
remarkable, and indicative of the simultaneous action
of a great catastrophe, that all the crater-margins are
either broken down or entirely destroyed on their
southern side. Part of what in older descriptions is
termed tufa consists of beds of palagonite, quite similar
to that of Iceland and Italy, as has been established
by Bunsen's exact analysis of the tufas of Chatham
island. (529) This, which is the easternmost island of
the entire group, and of which the exact position has
been astronomically determined by Beechey, is still 536
geographical miles distant from Punta de S. Francisco
on the American continent, according to my determina-
tion of the longitude of the town of Quito, 81° 4' 38" W.
from Paris, and Acosta's Mapa de la Nueva Granada of
1849.

ON ITS EXTEKIOR. VOLCANOES. 387

9. Mexico.

The six Mexican volcanoes, Turtla*, Orizaba, Popo-
catepetl*, Toluca, Jorullo*, and Colima*, four of which
have been burning within historic times, have already
been enumerated (53°), and their geologically remarkable
relative positions described. According to recent ex-
amination by Grustav Rose the formation of the rock of
Popocatepetl, the great volcano of Mexico, is the same
as that of Chimborazo, consisting, like it, of oligoclase
and augite. Even in the almost black beds of trachyte,
resembling pitchstone, oligoclase can be recognised in
very small crystals with oblique angles. The volcano of
Colima on the western side, on the shore of the Pacific,
belongs to the same Chimborazo and Teneriffe formation.
I have not myself seen Colima; but we owe to Pie-
schel(531) (since the spring of 1855) a very instructive
review of the rocks collected by him, as well as interest-
ing geological notices upon the volcanoes of the entire
Mexican highland, all of which he has himself visited.
The volcano of Toluca, whose narrow highest summit of
difficult access (the Pico del Frayle), was ascended by
me on the 29th of September, 1803, and its height de-
termined barometrically at 15,168 feet, is very different
in mineralogical composition from the still active Popo-
catepetl, and the volcano of Colima, which is not to be
confounded with another higher, snow-clad summit of
the same name. The volcano of Toluca consists, like
the Peak of Orizaba, Puy de Chaumont in Auvergne,
and JEgina, of an association of oligoclase and horn-
blende. According to the above short statement we
perceive that, in the long series of volcanoes extending
cc 2

388 REACTION OF THE INTERIOR OF THE EARTH

from sea to sea;, no two successive members have the
same mineralogical composition, a circumstance very
deserving of attention.

North-western America
(North of the parallel of the Rio Gila).

In the section which treated of the volcanic activity
of the East- Asiatic islands (332), particular importance
was attached to the bow-shaped curvature of the fissure,
of elevation from which the Aleutian islands have risen,
and which manifests an immediate connection between
the Asiatic and American continents, i. e. between the
two volcanic peninsulas Kamtschatka and Aliaska
(Alashka). We may call this the northern boundary of
a great bay of the Pacific Ocean, which, from a breadth
of 150° of longitude which it occupies at the equator,
diminishes to 3 7° of longitude between the points of the
two above-named peninsulas. On the American conti-
nent, near the sea, a number of more or less active volca-
noes have been known to mariners for the last seventy
or eighty years; but this group has been hitherto
regarded as isolated, unconnected either with the vol-
canic series of tropical Mexico, or with the volcanoes
supposed to exist in the peninsula of California. Now,
when a series of extinct trachytic cones are regarded as
intermediate links, an insight into their important geo-
logical connection has been gained by the filling up of
what was previously a gap, extending over more than
28° of latitude> between Durango and the new Wash-
ington territory north of West Oregon. This important
step in physical geography is due to the expedition sent

ON ITS EXTERIOR. VOLCANOES. 389

by the United States government to seek for the best
route from the plains of the Mississippi to the shores of
the Pacific ; the expedition was also well prepared in
scientific respects, and all parts of natural history have
benefited by it. In the now explored terra incognita
of this interval, great strips of country, very near to
the Eocky Mountains on their eastern side, and extending
to a considerable distance from their western declivities,
have been found covered by the products of either
extinct or still active volcanoes. We may thus see that,
beginning from New Zealand, and proceeding first for a
considerable distance in a north-west direction we can
pass through New Gruinea, the Sunda islands, the
Philippines, and the east of Asia, and ascending to the
Aleutian islands, can redescend to the. southward
through the north-western part of America, Mexico,
Central and South America, to the extremity of Chili ;
thus making the entire circuit of the Pacific Ocean, and
finding it surrounded, throughout a length of 26,400
geographical miles, by a series of recognisable monu-
ments of volcanic activity. Such a general cosmical
view could not have been gained, and based on adequate
grounds, without entering into particulars of exact geo-
graphical direction and improved classification.

Of the circumference of the great sea-basin here spoken
of, (or, inasmuch as there is but one throughout-com-
municating body of oceanic waters upon the earth, we
ought rather to say (533), of the circumference of the
largest of those portions of the ocean which form as it
were bays or gulfs between continents,) there still re-
main to be described the regions which extend from
the Eio Grila to Norton's and Kotzebue's Sounds. Analo-

390 KEACTION OF THE INTERIOR OF THE EARTH

gies derived from Europe, from the chains of the
Pyrenees or the Alps, and from South America from
the Cordilleras of the Andes from the south of Chili to
the fifth degree of north latitude in New Granada, sup-
ported by fanciful representations in maps, have propa-
gated the erroneous opinion that the Mexican high
mountains, or at least their highest ridge, can be traced,
as a wall-like rampart, from south-east to north-west
under the name of a " Sierra Madre." In reality, how-
ever, the mountainous part of Mexico is a broad mighty
intumescence, which does indeed hold its way continu-
ously at a height of from 5300 to 7400 feet, in the
assigned direction, between the two seas ; but upon
which, as in the Caucasus and in Central Asia, loftier
volcanic mountain-systems following partial and very
different directions rise to above 15,000 and 17,800 feet.
The directions of these partial groups, which have broken
forth over fissures which are also not parallel with each
other, are for the most part independent of the ideal
axis which can be drawn through the middle of the
whole swelling wave of the flattened ridge. These
remarkable relations in the form of the ground, occa-
sion an illusion which heightens the picturesque effect of
this beautiful land. The giant mountains clothed with
perpetual snow appear to rise as from a plain. The
surface of the softly swelling undulation, or the " high
plain," is scarcely distinguished from the plains of the
lowlands; and it is only the climate, the diminished
temperature under the same parallel of latitude, which
reminds us that we have ascended. The often-men-
tioned elevation-fissure of the volcanoes of Anahuac
(running east and west between the parallels of 19°

ON ITS EXTERIOR. VOLCANOES. 391

and 19 %° N. lat.) cuts the axis of the general undulation
almost at right angles. (534)

The form of a considerable part of the earth's sur-
face here indicated, and which was only begun to be
made out by careful measurement since 1803, is not to
be confounded with swellings of the surface which are
found included between two wall-like bounding moun-
tain chains, as in Bolivia around the lake of Titicaca,
and in Central Asia between the Himalaya and the
Kuen-lun. The first of these two last-named swellings,
the South American, which at the same time forms as
it were the floor of the valley, has, according to Pent-
land, a mean elevation of about 12,850 feet above the
sea; and the Thibetian, according to Captain Henry
Strachey, Joseph Hooker, and Thomas Thomson, an
elevation of about 15,000 feet. The wish expressed by
me half a century ago, in my very detailed Analyse de
1' Atlas geographique et physique du royaume de la
Nouvelle Espagne (§ xiv.), that my profile of the high
plain between the city of Mexico and Guanaxuato might
be continued by measurements over Durango and Chi-
huahua to Santa Fe del Nuevo Mexico, has now been
completely fulfilled. The distance, allowing only one
fourth for the inflections, is much more than 1200 geo-
graphical miles ; and the peculiar characteristics of this
long unregarded feature of the earth's surface (the
gentle slope and great breadth of the undulation, i. e.
from 240 to 280 geographical miles) are evidenced by
the circumstance that the distance between Santa Fe
and Mexico, (a difference of latitude of fully 16° 20',
and a distance about equal to that between Stockholm
and Florence,) can be travelled, on the table land, in

392 REACTION OP THE INTERIOR OF THE EARTH

four-wheeled carriages, without artificial roads. The
possibility of such communication was already known
to the Spaniards at the end of the sixteenth century,
when the viceroy, the Conde de Monterey (535), planned
the first settlement of Zacatecas.

In confirmation of what has been said above in gene-
ral of the relations of altitude between the city of
Mexico and the town of Santa Fe del Nuevo Mexico, I
subjoin the principal elements of the barometric level-
lings taken from 1803 to 1847. I have given the
places in the order from north to south, so that the
more northerly being highest in the page may corre-
spond most conveniently to the arrangement in ordinary
maps. (53S) (The initials indicate the observer, W. signi-
fies Wislizenus ; B., Burkart ; H., Humboldt.)

Santa Fe del Nuevo Mexico (lat. 35° 41' N.). Eleva-
tion 7046 feet, W.

Albuquerque (537) (35° 08' N.). Elevation 4849
feet, W.

Paso del Norte(538) on the Eio Grande del Norte
(29° 48' K). Elevation 3791 feet, W.

Chihuahua (28° 32' N.), 4638 feet, W.

Cosiquiriachi 6273, feet, W.

Mapimi in the Bolson de Mapimi (25° 54' K), 4782
feet, W.

Parras (25° 32' N.), 4985 feet, W.

Saltillo (25° IV N.), 5240 feet, W.

Durango (24° 25 ' N.), 6848 feet, Oteiza.

Fresnillo (23° 10' N.), 7244 feet, B.

Zacatecas (22° 50' K), 9012 feet, B.

San Luis Potosi (22° 08' N.), 6090 feet, B.

ON ITS EXTEEIOE. VOLCANOES. 393

Aguas calientes (21° 53' N.), 6262 feet, B.

Lagos (21° 20' N.), 6377 feet, B.

Villa de Leon (21° 07' N.), 6134 feet, B.

Silao, 5911 feet, B.

Guanaxuato (21° 0' 15" K), 6836 feet, H.

Salamanca (20° 40' N.), 5762 feet, H.

Celaya (20° 38' N.), 6017 feet, H.

Queretaro (20° 36' 39" N.), 6363 feet, H.

Sa'n Juan del Eio in the State of Queretaro (20° 30' K),

6491 feet, H.

Tula (19° 57' N.}, 6734 feet, H.
Pachuca, 8140 feet, H.
Moran, near Eeal del Monte, 8511 feet, H.
Huehuetoca, at the northern end of the great plain

of the city of Mexico (19° 48' N.), 7533 feet, H.
Mexico (19° 25' 45" N.), 7469 feet, H.
Toluca (19° 16' N.), 8824 feet, H.
Venta de Chalco, south-eastern end of the plain of

the city of Mexico (19° 16' N.), 7712 feet, H.
San Francisco Ocotlan, western end of the great plain

of Puebla, 7680 feet, H.
Cholula, at the foot of the ancient terraced pyramid

(1 9° 02' K), 6906 feet, H.

La Puebla de los Angeles (19° 0' 15"K), 7200 feet, H.
The village of Las Vigas marks the eastern end of the

high plain of Anahuac, 19° 37' N. ; the elevation of

the village is 7814 feet, H.

Whereas before the commencement of the nineteenth
century not a single barometric measurement of altitude
had been made in New Spain, it has now become pos-
sible to assign, as above, a chain of thirty-two stations

394 REACTION OF THE INTERIOR OF THE EARTH

throughout a zone of more than 16° of latitude, from
Santa Fe to the city of Mexico, all hypsometrically
determined, and the greater part having also their geo-
graphical positions fixed astronomically. In looking
over this zone we see the surface of the high broad
Mexican plain undulate between 5500 and 7000 French
feet (5862 and 7460 English). The lowest portion of
the route from Parras to Albuquerque is still more
than a thousand feet higher than the highest part of
Vesuvius.

This great but gentle swelling (539), the highest part
of which we have now been considering, increases so
much in breadth from its tropical portion to the paral-
lels of 42° and 44°, that the breadth of the " Grreat
Basin," west of the Mormons' Salt Lake, is upwards of
340 geographical miles, with a mean elevation of 4260
feet. It is everywhere very distinct from the rampart-
like chains of mountains which surmount it. The
knowledge of this form is one of the principal fruits of
Fremont's great hypsometrical investigations in 1842
and 1844. The swelling belongs to a different and
earlier epoch than the comparatively late elevation of
the mountain-ranges and systems of varying direction.
At the point where, about the 32d parallel of latitude
according to the present boundary-determinations, the
mountains of Chihuahua enter the western territory of the
United States, (the provinces which have been separated
from Mexico,) the mountains have already assumed the
somewhat indefinite name of Sierra Madre. It is not,
however, until near Albuquerque that a decided bifur-
cation (54°) takes place. At this bifurcation the western
branch retains the general name of Sierra Madre, and

ON ITS EXTERIOR. VOLCANOES. 395

the eastern one, from 36° 10" lat. northwards (a little to
the north-east of Santa Fe), has received from American
and English travellers the now everywhere recognised,
though not happily selected, name of the Kocky Moun-
tains. The two chains form between them a longitudi-
nal valley in which Albuquerque, Santa Fe£ and Taos
are situated, and through which the Rio Grande del
Norte flows. In 38^° N. lat. the valley is closed by a
chain running east and west for nearly 90 geographical
miles. The Rocky Mountains are continued in an
undivided north and south line to 41° N. In this inter-
val rise, rather on the eastern side, the Spanish peaks;
Pike's Peak (5798 feet), of which Fremont has given a
fine view; James's Peak (11,433 feet); and the three
Park Mountains, which enclose three high caldron-
shaped valleys whose lateral walls rise in the eastern
Long's Peak, or Big Horn, to heights of 9060 and 11,190
feet. (541) At the eastern edge, the boundary between
Middle and North Park, the chain of mountains alters
its direction, and from 40£° to 44° N., for a distance of
about 260 geographical miles, runs from south-east to
north-west. It is in this portion that the South Pass,
7490 feet, occurs, as well as the wonderfully pointed
and serrated Wind-River Mountains, with Fremont's
Peak (43° 08') which rises to the height of 13,567 feet.
In the parallel of 44° N., near the Three Tetons, the
north-western direction ceases and the north and south
direction of the Rocky Mountains is resumed. It is
maintained as far as Lewis and Clarke's Pass in 47° 02'
N., 1 12° W. There the chain of the Rocky Mountains
has still a considerable height (5977 feet), but on ac-
count of the many deep river beds towards Flathead

"396 KEACTION OF THE INTERIOR OF THE EARTH

River (Clarke's Fork), it becomes more regular and
simple. Clarke's Fork and Lewis (or Snake) Eiver
form the great Columbia Eiver, which will one day be
-an important commercial route. (Explorations for a
railroad from the Mississippi River to the Pacific Ocean
made in 1853-1854, vol. i. p. 107.)

As is the case in Bolivia, where the eastern chain of
the Andes, in which are the mountains of Sorata (21,288
feet) and Illimani (21,148 feet), does not now present
any still active volcanoes, so also in these western parts
of the United States volcanic activity is at present
limited to the coast chain of California and Oregon.
The long chain of the Rocky Mountains, of which the
distance from the sea coast of the Pacific varies from
480 to 800 geographical miles, and having no traces of
still-subsisting volcanic activity presents, however, like
the eastern chain of Bolivia in the valley of Yucay (542),
on both its declivities volcanic rocks, extinguished
craters, and even lavas with included obsidian and fields
of scoriae. In this part of the Rocky Mountains, which
has been described from the excellent examinations of
Fremont, Emory, Abbot, Wislizenus, Dana, and Jules
Marcou, the last-named distinguished geologist enume-
rates three groups of ancient volcanic rocks on the two
declivities. The earliest proofs of the volcanicity of this
district were also due to Fremont's observational habits,
commencing in 1842 and 1843. (Report of the Ex-
ploring Expedition to the Rocky Mountains in 1842,
and to Oregon and North California in 1843 — 1844,
pp. 164, 184—187, and 193.)

On the eastern slope of the Rocky Mountains, on the
south-western route from Bent's Fort on the Arkansas

ON ITS EXTERIOR. VOLCANOES. 397

River to Santa Fe del Nuevo Mexico, there are two
extinct volcanoes, the Eaton Mountains (543), with
Fisher's Peak, and (between Gralisteo and Pena Blanca)
the hill of El Cerrito. The lavas of the first of these
cover over the whole country between the Upper Arkan-
sas and the Canadian River. The peperino and volcanic
scoriae, which begin already to be found in the prairies
in approaching the Rocky Mountains from the eastward,
may perhaps belong to ancient eruptions of the Cerrito,
or possibly even of the great Spanish peaks (37° 32' N.).
This eastern volcanic domain of the isolated Raton
Mountains forms an area of 80 geographical miles in
diameter ; its centre is somewhere about 36° 50' N.

On the western slope of the Rocky Mountains, the
most clearly marked evidences of ancient volcanic ac-
tivity occupy a much wider space, which has been
traversed throughout its breadth from east to west by
the important expedition of Lieutenant Whipple.
This area, which is exceedingly irregular in outline, and
is moreover interrupted for a length of fully 120 geo-
graphical miles by the Sierra de Mogoyon, is comprised
(always speaking according to Marcou's geological map)
between the parallels of 33° 48' and 35° 40'. These
outbursts have therefore been more to the south than
those of the Raton Mountains : their mean latitude is
nearly that of Albuquerque. The area thus spoken of
divides itself into two sections ; that of the part of the
crest of the Rocky Mountains near Mount Taylor,
which terminates in the Sierra de Zuni (544), and the
more western division, called Sierra de San Francisco.
The conical Mount Taylor, 12,256 feet high, is sur-
rounded by radiating lava-streams, which can be dis- i

398 REACTION OF THE INTERIOR OF THE EARTH

tinctly traced as " malpais," still devoid of vegetation,
and covered with scoriae and pumice, following a winding
course for distances of many miles, in the same manner
as around Hecla. About 70 geographical miles to the
west of the present Pueblo de Zuiii rises the lofty
volcanic mountain ridge of San Francisco itself, whose
summit is estimated at full 16,000 feet; it runs to the
south of the Rio Colorado Chiquito, and further to the
west follow Bill William Mountain, the Aztec Pass
(6279 feet), and the Aquarius Mountains (8526 feet).
The volcanic rock does not terminate at the confluence
of the Bill William Fork with the Great Colorado River,
near the village of the Mohave Indians (in 34J°N.
114° W.); for several extinct, but stil] open, craters of
eruption can be recognised beyond the Rio Colorado,
near the Soda Lake. (545) Thus we see here, in the
present New Mexico, in the volcanic group of the Sierra
de San Francisco, to a little to the westward of the Rio
Colorado Grande or del Occidente (into which the Gila
falls), throughout an extent of 180 geographical miles,
a repetition of the ancient volcanic domain of Auvergne
and the Vivarais, and the opening of a new and wide
field to geological research.

Also on the western side, but 540 geographical miles
more to the north, we find the third ancient volcanic
group of the Rocky Mountains, that of Fremont's Peak
and of the two " Three Mountains," which are very much
like the Trois Tetons and the Three Buttes (546), both in
conical form and in name : they are to the west of these,
and therefore further from the chain. They present
wide-spread and much rent and torn black banks of
lava, with a scoriaceous surface. (547)

ON ITS EXTERIOR. TOLCANOES. 399

Several littoral or coast chains run parallel to the
chain of the Rocky Mountains, sometimes in single,
sometimes in double ranges ; and in their northern por-
tions (from 46° 12' N.) are still the seat of volcanic
activity. We have first, from San Diego to Monterey
(from 32i°to 36j°N.), the "Coast Range," specially so
called, a continuation of the ridge of land of the penin-
sula of Old or Lower California ; then, generally at a
distance of about 80 geographical miles from the shore
of the Pacific, the Sierra Nevada (de Alta California),
from 36° to 40 J°; — then, beginning from the high
Shasty Mountains in the parallel of Trinidad Bay
(41° 10'), the Cascade Range, which includes the highest-
summits still burning, and extends from south to north,
at a distance of 104 miles from the coast, to far beyond
the parallel of the Fuca Strait. Running in the same
direction, from 43° to 46° N., but at a distance of 280
miles from the coast, the Blue Mountains rise to a mean
height of 8000 feet. (548) In the middle portion of Old
California, but a little to the north, near the eastern
coast (the western shore of the gulf), in the district of
the former mission de San Ignacio, in about 28° N., we
find the extinct volcano, called the "Volcanes de las
Virgenes," which I have set down in my map of Mexico.
This volcano had its latest eruption in 1746 : we require
more certain information respecting both it and the
whole surrounding district. (See Venegas, Noticia de
la California, 1757, t. i. p. 27 ; and Duflot de Mofras,
Exploration de 1'Oregon et de la Californie, 1844, t. i.
pp. 218 and 239.)

Rocks belonging to ancient volcanoes have been already
found in the Coast Range near the port of San Francisco,

400 REACTION OF THE INTEMOK OF THE EAETII

at the Monte del Diablo (3672 feet), which has been ex-
amined by Dr. Trask, in the gold-yielding longitudinal
valley of the Eio del Sacramento, and in a fallen-in
trachytic crater called the Sacramento Butt, which has
been drawn by Dana. Further to the north, the Shasty
or Tshashtl Mountains, contain basaltic lavas ; obsidian,
used by the natives for arrow-points ; and the talcose
serpentines, which at many points of the earth's surface
occur in near connection with volcanic formations. But
the proper seat of still-subsisting volcanic activity is
found in the Cascade Mountains, where several peaks,
covered with perpetual snow, rise to heights of 15,000
and 16,000 feet. I subjoin a list of such moun-
tains, proceeding from south to north, and, as before,
marking by an asterisk those which are still more or
less active volcanoes. The lofty conical mountains not
so marked are probably in part extinct volcanoes, and in
part unopened trachytic summits.

Mount Pitt or M'Laughlin, in 42° 30', rather to the
west of Lake Tlamat, height 9549 feet.

Mount Jefferson or Vancouver, in 44° 35' ; a conical
mountain.

Mount Hood, in 45° 10' ; certainly an extinct volcano,
covered with cellular lava: according to Dana, it
and Mount St. Helen's are between 15,000 and
16,000 feet high, Mount St. Helen's being a little
the highest. (549) Mount Hood was ascended in
August 1853, by Lake, Travaillot, and Heller.

MountSwalalahos, or Saddle Hill, S.S.E.of Astoria (55°),
with a fallen-in extinct crater.

Mount St. Helen's *, north of the Columbia River, in

ON ITS EXTERIOR. VOLCANOES. 401

46° 12' K, at least upwards of 15,000 feet high,
according to Dana (551) ; still burning, always send-
ing forth smoke from the summit-crater : a volcano
covered with perpetual snow, of very fine and
regular conical form: on the 23d of November,
1842, a great eruption took place, which, according
to Fremont, has covered the country all around for
a great distance with ashes and pumice.

Mount Adams, in 46° 18' K, almost due east of Mount
St. Helen's; more than 112 geographical miles
from the coast : the last-named still-burning moun-
tain being only 76 miles therefrom.

Mount Reignier *, also written Mount Eainier, in
46° 48' N. ; E.S.E. from Fort Nisqually on Puget's
Sound, which is connected with the Fuca Strait;
a burning volcano : according to Edwin Johnson's
map in 1854, 12,330 feet high: it had violent
eruptions in 1841 and 1843

Mount Olympus, in 47° 50' K, only twenty-four
geographical miles south of the strait of San Juan
de Fuca, so long celebrated in the History of Dis-
covery in the Pacific.

Mount Baker*; a great and still active volcano in
the territory of Washington (in 48° 48'), rising to
a great height (? not yet measured), and of very
regular conical form.

Mount Brown (16,000 feet?), and a little to the east
of it Mount Hooker (16,730 feet?), are, according
to Johnson, lofty mountains of ancient volcanic
trachyte in New Caledonia, in 52 £° N. and 118°
and 120° W., remarkable therefore for being more
than 300 geographical miles from the coast.

VOL. IV. D D

402 REACTION OF THE INTERIOR OF THE EARTH

Mount Edgecombe*, on the small island called Lazarus
island, near Sitka, in 57° 3' N., whose violent fiery
eruption in 1796 I have already spoken of in the
present volume (English translation, Note 387).
Captain Lisiansky, who ascended it in the early
part of the present century, found the volcano not
then burning. Its height (552) according to Ernst
Hofmann is 3040 feet, according to Lisiansky 2801
feet: there are hot springs near it which break
forth from granite like those which are on the route
from the Valles de Aragua to Portocabello.

Mount Fair-weather, Cerro de Buen Tiempo, accord-
ing to Malaspina 14,710 feet high(553), in 58° 45' N.,
covered with pumice, was probably still burning in
recent times, like Mount Elias.

Volcano of Cook's Inlet (60° 08' K), according to
Admiral Wrangel 12,064 feet high; regarded by
that distinguished and accomplished navigator, as
well as by Vancouver, as an active volcano. (554)

Mount Elias, in 60° 17' N. and 136° 08' W.: accord-
ing to manuscripts of Malaspina found by me in
the archives of Mexico, 17,851 feet high; accord-
ing to the map of Captain Denham, 1853-1856, its
height is only 14,968 feet.

The phenomenon seen by the Arctic voyagers in a
recent expedition to the east of the mouth of Mackenzie
Kiver (in 69° 57' K and 127° W.), and which M'Clure
called " the Volcanoes of Franklin Bay," seems to have
been of the character either of what are sometimes
termed " Earth-fires," or of Salses emitting hot sul-

ON ITS EXTERIOR. VOLCANOES. 403

phurous vapours. An eye-witness, the missionary
Miertsching, interpreter to the expedition, found
thirty or forty columns of smoke which rose from
earth-fissures or from small conical elevations of
clays of many colours. The smell of sulphur was so
strong that one could hardly approach within twelve
paces of the columns of smoke. No rock in situ, or
solid masses, were found. Luminous phenomena were
seen at night from the ship ; no eruptions of mud were
observed, but great heat of the sea-bottom, and also
small basins of sulphuric acid and water. The district
deserves to be carefully examined. This phenomenon is
not to be connected with the volcanic activity of the
Californian Cascade mountains, the Cerro de Buen
Tiempo, or of Mount Elias. (M'Clure, Discovery of
the North-West Passage, p. 99 ; Papers relating to the
Arctic Expeditions, 1854, p. 34; Miertsching's Keise
Tagebuch, Grnadau 1855, S. 46.)

I have thus far described in their internal connection,
and in ascending order, the volcanic vital activities of
our planet ; the great and mysterious phenomenon of
the reaction of the molten interior against the surface
covered with vegetable and animal organic life. Next
in succession to the almost purely dynamical effects of
earthquakes (of the waves of commotion), I have placed
thermal springs and salses, i. e. phenomena which,

D D 2

404 REACTION OF THE INTERIOR OF THE EARTH

with or without spontaneous ignition, are produced by
permanently increased temperature imparted to the
waters of the springs and to gaseous exhalations, and
by chemical diversity of composition. The highest,
and, in its manifestations, the most complicated degree
of activity is presented in volcanoes, as these elicit the
great and diversified processes of cystalline rock-forma-
tion by the dry method, and are therefore not mere
agents of dissolution and destruction, but are also
formative, and bring substances into new combinations.
A considerable portion of very recent, if not of the
most recent, rocks are the work of volcanic activity;
whether, as is still the case at many points of the earth's
surface, by pouring forth molten masses from their own
conical or dome-shaped frameworks ; or whether, in the
youth of our planet, basaltic and trachytic rocks may
have flowed forth directly in a fluid state from a net-
work of open fissures, in proximity to sedimentary
strata.

I have endeavoured, in the preceding pages, to deter-
mine in the most careful manner the locality of the points
at which communication between the fluid interior of the
earth and the external atmosphere has long remained
open. I have now to sum up the number of these
points, to distinguish the volcanoes which still continue
active from among the much greater abundance of such
as have been so within remote historic periods, and to
consider them in their distribution as continental and as
insular volcanoes. If the entire number which I think
I may assume as the lowest (" nombre limite," " limite
inferieur ") were simultaneously in a state of activity,

ON ITS EXTERIOR. VOLCANOES. 405

the influence on the constitution of the atmosphere,
and on its climatic, and especially its electric, delations
would assuredly be very sensible; but the non-simul-
taneity of the eruptions diminishes the effect and
restricts it within very narrow, and for the most part
merely local limits. In great eruptions there arise
around the crater, in consequence of evaporation, vol-
canic storms, which, accompanied by lightning and
violent falls of rain, often occasion devastation ; but
an atmospheric phenomenon of this kind has no gene-
ral results. For, on account of the magnitude of .the
phsenomenon, I have always regarded as exceedingly
improbable the opinion which is even still sometimes
expressed, attributing the remarkable darkness which
for many months, from May to August in 1783, over-
spread a -considerable portion of Europe and Asia as
well as of the north of Africa (while on the high moun-
tains of Switzerland the sky was seen clear and unob-
scured), to great volcanic activity in Iceland, and to the
earthquake in Calabria. Where earthquakes embrace a
wide area, a certain influence on the period of the com-
mencement of the rainy season, as in the highlands of
Quito and Eiobamba in February 1797, and the south-
east of Europe and Asia Minor in the autumn of 1856,
may be more readily admitted, than extensive meteoro-
logical effects from an isolated volcanic eruption.

In the subjoined table the last column but one in-
dicates the number of volcanoes which have been cited
in the preceding pages; and the last column (within
brackets) indicates the portion of them which have
given proofs of activity within recent times.

406 REACTION OF THE INTERIOR OF THE EARTH

Number of Volcanoes on the Globe.

1. Europe . *

2. Islands of the Atlantic

3. Africa . V ';'.;

4. Continent of Asia

0. Western portion and Cen-

tral Asia
/8. Peninsula of Kamtschatka

5. East- Asiatic Islands . . "«'"

6. South-Asiatic Islands . •"*>

7. Indian Ocean . ; • . « .

8. Pacific Ocean . .,.. ^ .

9. Continent of America

a. South America.

a. Chili. . ; 4)

/8. Peru and Bolivia
y. Quito and New Gra-
nada . .

1. Central America . ;-• '.)•:

c. Mexico, south of the Rio

Gila . -V

d. North - west America,

north of the Gila

10. Antilles (555) (or West-India
Islands) . .

Pages.

326—328
328—332
332—334

7 (4)

14 (8)

3 (1)

334—342

11

(6)

342—348

14

(9)

349—361

69

(54)

361—367

120

(56)

367—371

9

(5)

372—386

40

(26)

273—278
273—277

24
14

(13)
(3)

273—276
262—267

18
29

(10)
(18)

267—269

6

(4)

388—405

24

(5)

note 555

5

(8)

Totals - 407 (225)

The result of this laborious investigation, which has
occupied me long because I have everywhere gone back
to the original authorities (the geological and geogra-
phical accounts of travellers), is that out of 407 vol-
canoes which have been cited, 225 have been in activity
within very modern times. Earlier statements (55G)
have given the number of still active volcanoes thirty

ON ITS EXTERIOR. VOLCANOES. 407

or fifty less, because prepared on different principles.
I have here restricted myself to volcanoes which either
still emit vapours, or which have had historically assured
eruptions within the nineteenth or the latter half of
the eighteenth century. There have indeed been in-
tervals between successive eruptions of the same volcano
extending over four centuries and upwards; but such
cases are very rare. We know the slow succession of
great eruptions of Vesuvius, in the years 79, 203, 512,
652, 983, 1138, and 1500. Previous to the great
eruption of Epomeo in Ischia in 1302, we know only
of those of the years 36 and 45 B.C., therefore fifty-
five years before the breaking out of Vesuvius.

Strabo, who died under Tiberius at the age of ninety,
(ninety-nine years after Spartacus had intrenched him-
self on Vesuvius), and who had no historic knowledge
of an older eruption, declares Vesuvius to be an
ancient volcano which had long ago burnt out and
become extinct. He says : " Above those places "
(Herculaneum and Pompeii) " lies Mount Vesuios,
encompassed by the finest cultivated fields, except on
its summit. This is, indeed, mostly a plain surface,
but unfruitful, and has an ashy appearance. It shows
cleft hollows of sooty-looking rock, appearing as if it
had been eaten into by the action of fire, so that we
may conjecture that at some former time there burnt in
these orifices a fire which has now become extinct, the
fuel which supported it having been consumed." (Strabo,
lib. v. page 247, Casaub.) This description does not
indicate either a cone of ashes or a crater-like depres-
sion (557) of the ancient summit, the encircling ridge of

408 REACTION OF THE INTERIOR OF THE EARTH

which might have served Spartacus (55S) and his gla-
diators as a rampart.

Diodorus Siculus (lib. iv. cap. 21, 5), who lived under
Caesar and Augustus, in describing the battle of Her-
cules with the Giants in the Phlegrsean Fields, also
designates " the Mount " now called Vesuvius, as a \6<f)o$,
which, like Etna in Sicily, once emitted much fire, and
(still) shows traces of ancient burning. He gives the name
of Phlegrsean Fields to the entire space between Cumae
and Neapolis. as does Polybius (lib. ii. cap. 17) to the still
larger space between Capua and Nola ; while Strabo (lib.
v.p. 246) describes the district near Puteoli (Dicsearchia),
where the great Solfatara is situated, with great local
truth, and terms it 'H^at'orou wyopd. In later times
the name ra ^>\s^pata TrsBla has been commonly limited
to this district, as in the present day geologists still
place the mineralogical composition of the lavas of the
Phlegrsean Fields in opposition to those of the district
round Vesuvius. We find the same opinion, that in
former times there had been fire under Vesuvius, and
that that mountain had anciently had eruptions, in
Vitruvius's book on Architecture (lib. ii. cap. 6), and
expressed in the most decided manner in a passage
which has not been sufficiently regarded. " Non minus
etiam memoratur antiquitus crevisse ardores et abunda-
visse sub Vesuvio Monte, et inde evomuisse circa agros
flammam. Ideoque nunc qui spongia sive pumex
Pompejanus vocatur, excoctus ex alio gen£re lapidis, in
hanc redactus esse videtur generis qualitatem. Id
autem genus spongiae, quod inde eximitur, non in
omnibus locis nascitur nisi circum ^Etnam et collibus
Mysiae qui a Graecis KaraKsicav^si'oi nominantur." As,

ON ITS EXTERIOK. VOLCANOES. 409

according to the investigations of Bockh and Hirt, there
can no longer remain any doubt as to Vitruvius having
lived under Augustus (559), therefore fully a century
before the eruption of Vesuvius in which the elder
Pliny perished, the passage which has been quoted
with the expression " pumex Pompeianus " (connecting
pumice with Pompeii) presents peculiar geological in-
terest in relation to the debated question; — whether, ac-
cording to Leopold von Buch's ingenious conjecture (56°),
Pompeii was covered by beds of tufa containing pu-
mice, of submarine formation, spread in horizontal strata
over the whole surface between the Apennine moun-
tains and the western coast from Capua to Sorrento
and from Nola to beyond Naples, and pushed up at
the first formation of the Somma; — or whether Vesu-
vius itself, quite contrary to its present habits, sent
forth the pumice from its own interior ?

Carmine Lippi (561), who (1816) ascribes the tufa
which overspread Pompeii to water, as well as his
acute opponent, Archangelo Scacchi (562), in the letter
addressed by him to the Cavaliere Francesco Avellino
(1843), have called attention to the remarkable circum-
stance, that a portion of the pumices of Pompeii and
of the Somma enclose small pieces of lime retaining
their carbonic acid, which, if they were exposed to
great pressure in the course of an igneous formation, is
not very surprising. I have myself had the opportunity
of seeing specimens of these pumices in the interesting
geological collections of my learned friend and academi-
cal colleague, Dr. Ewald. This similarity of mineralo-
gical constitution at two opposite points must give
occasion to the question, whether the covering which

410 REACTION OF THE INTERIOR OF THE EARTH

overspread Pompeii was, as Leopold von Buch would
consider, thrown down from the declivities of the Somma
in the eruption of 79, — or whether, as Scacchi maintains,
the newly opened crater of Vesuvius cast forth pumice
simultaneously on Pompeii and on the Somma? The fact
of "pumex Pompeianus" being known to Vitruvius in the
time of Augustus, points to eruptions anterior to Pliny :
and from the experience which we have had of varia-
bility in the formations by volcanic activity in different
ages and under different circumstances, we are by no
means warranted either in absolutely denying the possi-
bility of Vesuvius having ever at any former period pro-
duced pumice, or in absolutely assuming that pumice,
i. e. a fibrous or porous state of a pyrogenous mineral, can
only be formed where obsidian or trachyte is present
with glassy felspar (sanidine).

Although, according to the examples adduced, much
uncertainty still subsists as to the length of the periods
after, which a slumbering volcano may re-awaken, yet it
is of great importance to ascertain the geographical
distribution of burning volcanoes at a determinate
epoch. Of the 225 orifices through which, in the
middle of the nineteenth century, the molten interior of
the earth is in volcanic communication with the atmo-
sphere, 70 (less than a third part therefore) are on con-
tinents, and 155 (or fully two thirds) are on islands. Of
the 70 continental volcanoes, 53 (or three fourths) belong
to America, 15 to Asia, 1 to Europe, and 1 or 2 to the
portion of Africa with which we are acquainted. It is
in the South-Asiatic Islands (the Sunda isles and the
Moluccas) and in the Aleutian and Kurile (East- Asiatic)
islands, that the greatest number of island volcanoes
are congregated within the smallest space. The Aleu-

ON ITS EXTERIOR. VOLCANOES. 411

tian islands contain perhaps more volcanoes active
within recent historic times than does the whole conti-
nent of South America. Taking the earth altogether,
it is the region comprised between the 73rd W. and
127th E. meridian from Greenwich, and the parallels
of 47° S. and 66° K, extending from south-east to
north-west in the more western part of the Pacific, which
is the richest in volcanoes.

If, in looking in a similarly general manner at the
great sea-gulf which we are accustomed to call the
Pacific Ocean, we consider it to be bounded on the north
by the parallel of Bering Strait, and on the south by
the parallel of New Zealand which is also that of South
Chili and North Patagonia, we find, — and it is a result
very deserving of notice, — that within this basin, and
on the American and Asiatic shores which surround it,
there are 198, or nearly seven eighths of the whole
number of active volcanoes (225) on the surface of the
globe. The volcanoes nearest to the poles are, — ac-
cording to our present geographical knowledge, — in the
northern hemisphere, the volcano of Esk, on the little
island of Jan Mayen, in 71° I' lat, and 7° 29' W. long.,
and, in the southern hemisphere, Mount Erebus, which
sends forth reddish flames visible even in daylight, and
which in 1841 was discovered by Sir James Eoss in his
great antarctic voyage (563), and found by him to be
12,400 feet high, or about 240 feet higher than the
Peak of Teneriffe, in 77° 33' lat. and 167° E. longitude
from Greenwich.

The great comparative frequency of volcanoes on islands
and on the coasts of continents, could not but early lead
geologists to inquire into the causes of this phenomenon.

412 REACTION OF THE INTERIOR OF THE EARTH

I have already referred, in an earlier volume (Kosmos,
Bd. i. S. 454, Engl. ed. vol. i. note 234), to the com-
plicated theory of Trogus Pompeius, under Augustus,
according to which the salt water of the sea is supposed
to stimulate the volcanic fire. Different chemical and
mechanical causes of the influence of the neighbourhood
of the sea have been adduced up to the most recent times.
The ancient hypothesis of the penetration of sea-water
to the volcanic hearth seemed to have acquired a more
solid foundation at the epoch of Davy's discovery of the
metallic bases of the earths; but the great discoverer
himself soon gave up that hypothesis, to which even Gray-
Lussac had been inclined (564) notwithstanding the rarity
or entire absence of hydrogen gas. Mechanical, or rather
dynamical causes, whether they should be looked for in
the folding of the earth's outer crust and the upheaval
of continents, or in the locally lesser thickness of that
crust, would seem, according to my ideas, to present a
greater degree of probability. We may readily repre-
sent to ourselves the probability that, at the margins of
the upheaving continents, whose coasts now rise with
more or less abruptness above the waters of the sea,
simultaneously occasioned subsidence of the ocean-bed
might cause the formation of fissures tending to pro-
mote communication with the molten interior. In the
inland parts of elevated continents, at a distance from
the oceanic areas of subsidence, there would not be the
same occasion of fracture. Volcanoes follow the coast-
lines in single, sometimes in double, and even triple
ranges. Short cross ridges, elevated over cross fissures,
connect these ranges, forming mountain-knots. Fre-
quently, but by no means invariably, it is the outer
range, nearest to the sea-shore, which is the most

ON ITS EXTEEIOE. VOLCANOES. 413

active, while the more inland ones are extinct or
appear approaching extinction. Sometimes persons
have thought that they could recognise within one
and the same range of volcanoes, an increase or de-
crease of eruptive activity in a definite direction, but
the phenomena which are sometimes seen of a re-
awakening after long periods of tranquillity render any
such recognition very insecure.

As, owing either to the absence or disregard of well-
assured determinations of position, both of the volcanoes
themselves and of the nearest points on the sea-shore,
there have been many inexact statements respecting the
distance of phenomena of volcanic activity from the
sea, I give here the following statements in geographical
miles (sixty to an equatorial degree). In the Cordil-
leras of Quito, Sangay, with its unintermitting activity,
is on the easternmost and most distant range, but yet it
is only 112 geographical miles from the sea. Very well
educated monks of the missions to the Andaquies
Indians on the Alto Putumayo have assured me, that
on the Upper Eio de la Fragua a tributary of the
Caqueta east of the Ceja, they have seen smoke issue
(565) from a not very high conical mountain, of which
the estimated distance from the sea was 160 miles. The
Mexican volcano of Jorullo, upheaved in September
1759, is eighty-four miles from the nearest point of the
sea-coast (Kosmos, Bd. iv. S. 339 — 346, present volume,
pp. 294 — 302); the volcano of Popocatepetl is 132
miles from the sea ; an extinct volcano in the eastern
cordillera of Bolivia, near S. Pedro de Cacha, in the
valley of Yucay (present volume, p. 277), is above 180
miles ; the volcanoes of the Siebengebirge near Bonn,

414 REACTION OF THE INTERIOR OF THE EARTH

and of the Eifel (Kosmos, Bd. iv. S. 275—282, pre-
sent volume, pp. 229—236) from 132 to 152 miles;
those of Auvergne, the Velay, and the Vivarais (566),
according to a division into three distinct groups (group
of the Puy de Dome near Clermont and the Mont-
Dore; group of the Cantal; and group of the Puy
and Mezenc), are respectively 148, 116, and 84 miles
from the nearest sea. South of the Pyrenees, the ex-
tinct volcanoes of Olot, to the west of Grerona, with
their very distinct, sometimes divided, lava-streams, are
only twenty-eight miles from the Catalonian coast of
the Mediterranean ; on the other hand, the undoubted,
and, according to all appearance only very recently ex-
tinguished, volcanoes in the long chain of the Rocky
Mountains in North-west America are from 600 to 680
miles from the coast of the Pacific.

A very abnormal phenomenon in the geographical
distribution of volcanoes is the existence of some which
have been active within historic times, and some of
which may perhaps be even still burning, in the chain
of the Thian-schan (the Celestial Mountains) in Central
Asia, between the two parallel chains of the Altai and
the Kuen-lun. Their existence was first made known
in Europe by Abel Eemusat and Klaproth ; and by the
aid of the sagacious and laborious sinological investiga-
tions of Stanislas Julien, I have been enabled to treat
of them more fully in my Asie Centrale. (567) The
distances of the volcano Pe-schan (Mont Blanc) with
its lava-streams, and the still burning Mount (Ho-tscheu)
of Turfan, from the shores of the Polar Sea and of the
Indian Ocean, are almost equal ; being about 1480 and
1520 miles. On the other hand, the volcano of Pe-

ON ITS EXTERIOE. VOLCANOES. 415

schan (whose lava-yielding eruptions, from the year 89
to the commencement of the seventh century, have
been severally recorded in Chinese works) is only 172
geographical miles from the great Alpine Lake Issikul,
situated on the south side of the Temurtutagh (a western
portion of the Thian-schan), and 208 miles from Lake
Balkasch which is 148 miles long. (568) The great Dsai-
sang Lake (near which I found myself in 1829, in
Chinese Dsungary) is 360 miles from the volcanoes of
the Thian-schan. There is, therefore, no deficiency of
inland waters, although, indeed, not in such near prox-
imity as that of the still active volcano of Demavend
(in Persian Mazanderan) to the Caspian Sea.

If we admit that masses of water, oceanic or in-
land, are not requisite for the maintenance of volcanic
activity, and that, as I am inclined to believe, islands
and coasts are richer in volcanoes only because the
upheaval effected by internal elastic forces is accom-
panied by adjacent depression in the bed of the sea(569),
— so that an area of elevation borders on an area of sub-
sidence, and at the limit between these areas great and
profound clefts and fissures are occasioned, — then we
may conjecture that, in the zone of Central Asia which
is included between the parallels of 41° and 48° K, the
great Aralo-Caspian basin of depression, as well as the
considerable number of lakes (some arranged in chains
and others not so arranged), may have occasioned, between
the Thian-schan and the Altai Kurtschum, the existence
of phenomena similar to those which elsewhere are found
in proximity to the sea-coast. We know by tradition that
many comparatively small pieces of water which now
form chains of lakes (like beads on a necklace, lacs a

416 REACTION OF THE INTERIOR OF THE EARTH

chapelet) were. once part of a single large basin. We
still see larger lakes become subdivided into smaller
ones, by reason of the existing disproportion between
the amount of water falling in the shape of rain or
snow and the amount withdrawn by evaporation. An
observer well acquainted with the Kirghis steppe,
General Genz in Orenburg, thought that there once
existed a water-communication between the Sea of Aral,
the Aksakal, Sary-Kupa, and Tschagli. We can re-
cognise a great furrow, running from south-west to
north-east, traceable beyond Omsk between the Irtysch
and the Obi, through the steppe of Barabinski with
its numerous lakes, towards the moory plains of the
Samoieds, Beresow, and the shore of the Icy Sea.
We may perhaps connect with this furrow the old and
widely prevalent tradition of a "Bitter Sea" (also
called the "Dried-up Sea," Hanhai), which extended
eastward and southward from Kami, and in which a
part of Gobi, of which the salt and reedy centre has
been recently found by the exact barometric measure
ments of Dr. von Bunge to be only 2558 feet above the
level of the sea, rose as an island. (57°) Seals, quite
similar to those which inhabit the Caspian and the
Baikal in troops, are found (and this is a geological fact
which has not yet received the attention it deserves) in
the little fresh-water lake of Oron, situated more than
400 geographical miles east of Lake Baikal. The lake
of Oron, which is only some miles in circumference, is
connected with the river Witim, a tributary of the
Lena, in which no seals live.(571) The present isolated
position of these animals, their distance from the mouth
of the Volga (fully 3600 geographical miles), is a very

ON ITS EXTERIOR. VOLCANOES. 417

remarkable geological phenomenon, for it indicates a
former water-communication of very extensive character.
May we suppose that the many and various subsidences,
to which this central part of Asia has been exposed
over a wide extent, may have exceptionally given occa-
sion, on the convexity of the continental swelling, to
circumstances and relations similar to those at the
borders of elevation-fissures on coasts ?

Well-assured reports made to the Emperor Kanghi
have made known the existence of a now extinct volcano
far to the east, in the north-west part of Mantchou
Tartary, in the district about Mergen, which is probably
in 48J° N. lat. and 122 E. long, from Greenwich. The
eruption of scoriae and lava from this Mount Bo-schan,
or Ujun-Holdongi (the Nine Hills), about fourteen miles
south-west of Mergen, took place in January 1721.
The hills of scoriae which were thrown up are said, by
the persons whom the Emperor Kanghi sent to examine
the effects, to have occupied a space of twenty-four
miles in circumference, and the same authorities stated
that a current of lava, by damming up the river Udelin,
occasioned the formation of a lake. From less circum-
stantial Chinese accounts, it would appear that Bo-schan
had had an earlier igneous eruption in the seventh
century. The distance from the sea is about 420 geo-
graphical miles ; more than three times the distance of
the volcano of Jorullo from the sea, and similar to that
of the Himalaya, (572) We owe these interesting notices
to W. P. Wassiljew (G-eograpK Bote, 1855, Heft 5, S.
31); and to a paper by Semonow (the learned translator
of Carl Eitter's Erdkunde), in the 17th volume of the
Memoirs of the Kussian Imperial G-eographical Society.

TOL. IV. E E

418 REACTION OF THE INTERIOR OF THE EARTH

In inquiries respecting the geographical distribution
of volcanoes, and their greater frequency on islands and
coasts (i. e. margins of elevation of continents), the
probable great inequality of the already attained thick-
ness of the earth's crust has also been brought into
discussion. One is inclined to assume the surface of
the interior molten mass to be nearer the points where
volcanoes have broken forth. But inasmuch as we
can imagine many intermediate degrees of consistency
in the solidifying materials, it is difficult to form
any clear conception of such a supposed surface of
the molten mass, if we are to look for the principal
cause of all overthrowals, fissurings, elevations, and
basin-shaped subsidences, in an alteration of capacity in
the external already solidified shelL If it were admissi-
ble to determine the so-called thickness of the solid
crust of the globe (573) from the data we have obtained
in Artesian wells and the temperature of fusion of
granite, by an arithmetical series, i. e. by assuming
throughout equal geothermic gradations of depth, we
should find it not quite twenty-one geographical miles,
or 3-^-9 th of the polar diameter (574) ; but the effects of
pressure and of conduction of heat in different kinds of
rock give us reason to suppose that the increment of
heat in descending will become less rapid as the depth
increases.

Notwithstanding the very small number of points at
which the molten interior of our planet is at the present
time in active communication with the atmosphere, it is
yet a question not without importance if we ask, what
is the kind and the measure of influence exercised by
volcanic gaseous exhalations on the chemical compo-

ON ITS EXTERIOR. VOLCANOES. 419

sition of the air, and through its medium upon the
development of organic life upon the Earth's surface ?
We must first bring under consideration, in reference to
this question, the circumstance that gases are exhaled
less from summit-craters themselves, than from small
cones of eruption and from the fumaroles which over
wide spaces surround so many volcanoes ; and that even
entire districts, in Iceland, in the Caucasus, in the
highlands of Armenia, in Java, the Gralapagos, the
Sandwich Islands, and New Zealand, manifest uninter-
rupted activity through solfataras, naphtha-springs, and
salses. Volcanic districts which are now reckoned as
extinct are also to be regarded as sources of gas, and it
is probable that the silent operation of subterranean
decomposing and formative forces taking place in them
is, in respect to quantity, more productive than the
grander but more rare phenomena of volcanic eruptions ;
although, in regard to these also, it is to be remembered
that the " fields of lava " continue for years to send
forth visible and invisible vapours. If it should be
thought that the effects of these small chemical pro-
cesses may be disregarded, because the enormous volume
of the atmosphere, incessantly impelled to and fro by
currents, can be so little altered by the addition to its
primitive mixture of such minute fractions appearing
severally so unimportant (575), we should remember the
powerful influence exercised, according to the fine inves-
tigations of Percival, Saussure, Boussingault, and Liebig,
by three or four ten-thousandth parts of carbonic acid
gas in our atmosphere upon the existence of vegetable
organic life. According to Bunsen's fine investigations
on the volcanic gases, some fumaroles, in different stages

EE 2

420 REACTION OF THE INTERIOR OF THE EARTH

of activity, and under different local circumstances,
give out (for example, at Hecla) 0'81 to 0-83 of nitrogen,
and in the lava-streams of that mountain 0-78, with only
traces (O'Ol to 0'02) of carbonic acid gas ; other fumaroles
in Iceland, as near Krisuvik, give 0-86 to 0-87 of car-
bonic acid, with scarcely so much as 0*01 of nitrogen. (576)
In the same manner the important investigations into
the emanations of gas in Southern Italy and in Sicily, by
Charles St. Claire Deville and Bornemann, show great
amount of nitrogen (0'98) in the exhalations from a
cleft deep within the crater of Vulcano, but they also
show sulphuric acid vapours with a mixture of 74*7
parts of nitrogen and 18*5 oxygen; not differing much,
therefore, from the constitution of atmospheric air.
The gas which rises in the fountain of Acqua Santa (577)
near Catania is on the other hand pure nitrogen, as was
at the time of my American journey the gas of the
Volcancitos de Turbaco. (578)

Are we to suppose that the large quantity of nitrogen
which is .diffused by volcanic activity is solely that
which is brought to the volcanoes by meteoric water ?
or are ther in the interior deeply situated sources of
nitrogen ? It is also to be remembered that the air con-
tained in rain-water, instead of 0*79, as our common air,
has, according to my experiments, only 0-69 of nitrogen.
The latter is a source of enhanced fertility by the for-
mation of ammonia through the operation of electric
explosions, which are of almost daily occurrence within
the tropics. (579) The influence of nitrogen on vegeta-
tion is similar to that of the substratum of atmospheric
carbonic acid.

Boussingault, in his analysis of gases from volcanoes

ON ITS EXTEEIOR. VOLCANOES. 421

near the equator (Tolima, Purace, Pasto, Tuqueres, and
Cumbal), found, together with much aqueous vapour,
carbonic acid and sulphuretted hydrogen, but no mu-
riatic acid, no nitrogen, and no free hydrogen. (58°) The
influence which the interior of our planet still exercises
on the chemical composition of the atmosphere, by
withdrawing these substances to return them again
under other forms, is no doubt a very inconsiderable
part of the chemical revolutions which, in primeval
times, the atmosphere must have undergone at the
breaking forth of great mountain masses from open fis-
sures. The conjectures which have been formed respect-
ing the probably very large portion of carbonic acid gas
in the ancient atmosphere, have been strengthened by
the comparison of the thickness of coal deposits with
the very thin carboniferous stratum (only seven lines in
thickness), which, in the temperate zone, according to
Chevandier's calculations, our thickest forests would
supply to the ground in 100 years. (-1581)

In the infancy of geology, prior to Dolomieu's saga-
cious conjectures, the source of volcanic activity was
not placed beneath the oldest rocks, which were then
universally held to be granite and gneiss. Resting on
some feeble analogies of inflammability, it was long
believed that the source of volcanic eruptions, and of
the emanations of gas occasioned by them, and lasting
for many centuries, was to be sought for in the combus-
tible materials contained in the newer upper silurian
sedimentary strata. More general knowledge of the
earth's surface, more profound and better directed geo-
logical researches, and the beneficial influence which
the great advances of modern chemistry have exercised

422 REACTION OF THE INTERIOR OF THE EARTH

on geology, have taught us that the three great groups
of volcanic or eruptive rocks (trachytes, phonolites, and
basalts), which, regarded in great masses, are different
from each other in age, and most often occur very
widely apart, are, however, all three more recent in
their appearance than the plutonic granites, diorites,
and quartzose porphyries, and than all silurian, second-
ary, tertiary, and quartary (pleistocene) formations ; and
that they even often intersect loose beds of diluvial
gravel and ossiferous breccias. Rozet has made the
important remark, that a striking multiplicity (582) of
such intersections are found congregated together within
a small space in Auvergne; for if the great trachytic
mountain masses of the Cantal, Mont Dore, and Puy de
Dome break through granite itself, and sometimes (as
between Vic and Aurillac, and at the Griou de Mamon)
enclose large fragments of gneiss (583) and limestone, yet
we also see trachyte and basalt intersect as dikes, gneiss,
carboniferous rocks, tertiary and diluvial strata. Basalts
and phonolites, which are nearly allied to each other (as
is shown by the Bohemian Mittelgebirge and Auvergne),
are both later formations than the trachytes, which are
often intersected by dikes of basalt. (684) Phonolites are
older than basalts, in which they have never been found
forming dikes, whereas dikes of basalt often traverse
porphyritic schists (phonolites). In the Andes of Quito
I found the basalts locally separated from the prevailing
trachytes, occurring almost alone at the Eio Pisque and
in the valley of Gruaillabamba. (585)

As in the volcanic high plain of Quito everything is
covered over by trachytes, trachyte conglomerates, and
tufa, my most earnest endeavours were bent to discover

ON ITS EXTERIOR. VOLCANOES. 423

some point at which it might be clearly seen what was
the older rock upon which the mighty cones and bell-
shaped mountains rose, or, more properly speaking?
through which they had broken forth. I was so fortu-
nate as to find such a point, at a height of 9483 feet
above the Pacific, in the month of June 1802, when
attempting, from Eiobamba Nuevo, the ascent of Tun-
gurahua, on the side of the Chuchilla de Gruandisava. I
was proceeding from the pleasantly situated village of
Penipe, by the oscillating rope-bridge (puente de ma-
roma) over the Eio Puela, to the solitary Hacienda de
G-uansce (7929 feet), where, on the south-east, opposite
to the junction of the Eio Blanco with the Eio Chambo,
a magnificent colonnade of black, pitchy-looking trachyte
rises to view. The appearance recalls that of the ba-
saltic quarry at Unkel seen from a distance. On the
Chimborazo, rather above the little lake of Yana Cocha,
I saw a similar, loftier, but less regular columnar group
of trachyte. The columns to the south-east of Penipe
are, for the most part, five-sided, only 15 inches in
diameter, and often bent and diverging. At the foot of
these pitch-black Penipe trachytes (not far from the
mouth of the Eio Blanco), one sees a phenomenon very
unlooked for in this part of the Cordilleras, viz., green-
ish-white mica-slate, with interspersed garnets; and,
farther on (beyond the shallow little stream of Basca-
guan, by the Hacienda of Gruansce, near the bank of
ihe Eio Puela), probably dipping under the mica-slate,
granite of a middling-sized grain, with bright reddish
felspar, a little blackish-green mica, and much grayish-
white quartz. Hornblende is wanting. It is not syenite.
The trachytes of the volcano of Tungurahua, which in,

424 REACTION OF THE INTERIOR OF THE EARTH

their mineralogical constitution are like those of Chim-
borazo, i.e. consist of a mixture of oligoclase and
augite, have thus here broken through granite and
mica-slate. Farther to the south, a little to the east
of the route from Eiobamba Nuevo to Gruamote and
Ticsan, what were formerly called the primitive rocks,
mica-slate and gneiss, come everywhere into view in
the Cordillera, which has here turned away from the sea,
appearing about the foot of the colossal Altar de los
Collanes, of Cuvillan, and of the Paramo del Hatillo.
Before the arrival of the Spaniards, and even before the
sovereignty of the Incas extended so far to the north,
the natives are supposed to have worked some metal-
liferous beds near the volcano. A little to the south of
San Luis, many dikes of quartz may be observed tra-
versing a greenish clay-slate. Near Gruamote, at the
entrance of the grassy plain of Tiocara, we find great
masses of quartzite, very poor in mica, of remarkably
linear parallel structure, inclining regularly 70° to the
north. More to the south, near Ticsan, not far from
Alausi, the Cerro Cuello de Ticsan presents great masses
of sulphur in a bed of quartz, subordinate to the adja-
cent mica-slate. Such an extensive distribution of
quartz in the vicinity of trachytic volcanoes at first
sight excites some surprise. But my observations of
the superposition, or rather, of the breaking forth of
trachyte through mica-slate and granite at the foot of
Tungarahua (a phenomenon which is as rare in the
Cordilleras as it is frequent in Auvergne), have been
confirmed, 47 years later, by the excellent researches of
the French geologist, Sebastian Wisse, at Sangay. This
colossal volcano, 1343 feet higher than Mont Blanc,

ON ITS EXTERIOR. VOLCANOES. 425

having no lava-streams (Charles Deville considers the
equally active volcano of Stromboli to be similarly with-
out such), but which, at least since the year 1728, has
been in constant activity, erupting black and often
brightly-glowing rocks and stones, forms an island of
trachyte, of scarcely so much as 8 miles diameter (586),
in the midst of granite and gneiss. Very different rela-
tions of position are presented, as I have already re-
marked, in the volcanic district of the Eifel, in the Maars
(mine-funnels) sunk in the Devonian schists, as well as
in the lava-yielding volcanic frameworks of the long
ridges of the Mosenberg and Grerolstein. The surface
does not here betray what is concealed within. The
absence of trachyte, in volcanoes which were so active
many thousand years ago, is a still more striking phe-
nomenon. The augite-containing scoriae of the Mosen-
berg, which in part accompany the basaltic lava-stream,
have embedded in them small burnt pieces of schist,
not fragments of trachyte; there are no trachytes in
the vicinity. This last-named rock is only visible in
the Eifel district (587), quite apart and at a distance
from Maars and lava-yielding volcanoes, as in the Sell-
berg, near Quiddelbach, and in the ridge of Eeimerath.
The variety of formations broken through by volcanoes,
in rising through the upper crust of the earth, are
geologically no less important than are the substances
which they produce.

In the most distant parts of the earth the forms of
the rocky frameworks through which the volcanic acti-
vity manifests itself, or has striven to do so, liave
been far more carefully examined and represented in
their often very complicated diversity, than was the

426 REACTION OF THE INTERIOR OF THE EARTH

case in the last century, when the whole morphology of
volcanoes was limited to cones and domes or bell-shaped
mountains. We are now most satisfactorily acquainted
with the structure, hypsometry, and arrangement (with
what the ingenious Carl Friedrich Naumann calls the
" geotektonik ") (588) of many volcanoes, while we are
still in great uncertainty respecting the composition of
their rocks, and respecting the association of mineral
species which characterise their trachytes, and can be
recognised apart from the ground mass. Both these
kinds of knowledge, however, — the " morphologi-
cal" for the rocky frameworks, and the "oryctognos-
tical" for their composition, — are alike necessary for
forming a complete judgment in respect to volcanic acti-
vity; indeed, the latter kind, based on crystallisation
and chemical analysis, may be regarded as the more
geologically important of the two, on account of the
connection with plutonic rocks (quartzose-porphyry,
greenstone, and serpentine). All the little which we
suppose we know of what we call volcanic character in
the moon is, from the nature of the case, restricted
solely to form. (589)

If, as I hope, any interest attaches to what I have
here said respecting the classification of volcanic rocks,
or rather respecting the division of trachytes according
to their composition, the merit of the grouping belongs
entirely to my friend and Siberian travelling compa-
nion, Gustav Kose. By his own observations in the
open field over extensive regions, and by a happy com-
bination of knowledge in chemistry, crystallographic
mineralogy, and geology, he has been peculiarly well
qualified for promulgating new views respecting the

ON ITS EXTERIOR. VOLCANOES. 427

minerals, whose varied, yet often repeated association is
the result of volcanic activity. He has, partly out of
kindness to myself, and at my solicitation, particularly
since 1834, repeatedly examined the specimens which I
brought back with me from the volcanoes of New Gra-
nada, Los Pastos, Quito, and the high land of Mexico,
and has compared them with those brought from other
parts of the world, which are contained in the rich
mineralogical collection of Berlin. Leopold von Buch,
when he was in Paris, in 1810 — 1811, between his re-
turn from Norway and his departure for Teneriffe, at a
time when my collections had not been separated from
those of my companion, Aime Bonpland, had examined
them microscopically with persevering diligence. He
had also previously, while with Gray-Lussac in Rome, in
the summer of 1805, and afterwards in France, taken
into consideration what I had written down at the time
and on the spot, in my travelling journal, both respect-
ing particular volcanoes and in general (in July 1802)
" Sur 1'affinite entre les volcans et certains porphyres
depourvus de quartz."(590) I preserve, as a valued me-
morial, some pages with remarks on the volcanic pro-
ducts of Quito and Mexico, which that great geologist
wrote for my information more than 46 years ago.
Since, as I have elsewhere explained more circumstan-
tially (591), travellers can but carry forward the incom-
plete knowledge of their own time, and their observations
are liable to be deficient in many of the guiding ideas
and of the distinguishing marks which are the fruits of
progressive knowledge, a time may arrive when the
materials they have collected and arranged geographi-
cally will almost alone retain a value.

428 REACTION OF THE INTERIOR OF THE EARTH

If, as has often been done, it is desired to restrict the
name of trachyte (for the sake of conformity with its
earliest application to the rocks of Auvergne and of the
Siebengebirge near Bonn,) by applying it exclusively to
a volcanic rock containing felspar (particularly Werner's
glassy felspar and Kose's and Abich's sanidine), the
result is unfruitfully to rend asunder those intimate
links between different volcanic rocks which conduct
to higher geological views. Such a restriction might
justify the expression, " that Etna, rich in labradorite,
has no trachyte ; and, according to it, my own collec-
tions would be held to prove that " none of the almost
countless volcanoes of the Andes consist of trachyte,
and even that the mass of which they are composed
should 4be called Albite, and, therefore, inasmuch as at
that time (1835) all oligoclase was erroneously supposed
to be albite, that all volcanic rocks would have to be
designated by the general name of Andesite (consisting
of albite with little hornblende). "(592) The impressions
which I brought back from my travels of that which,
notwithstanding diversity of mineralogical composition,
all volcanoes possess in common, have been confirmed
by Grustav Rose, who, in accordance with the views and
evidence unfolded in his fine memoir on the felspathic
group (593), has, in his classification of trachytes, regarded
orthoclase, sanidine, the anorthite of the Somma, albite,
labradorite, and oligoclase, in a general point of view,
as the felspathic portion of volcanic rocks.

Short denominations, purporting to contain defini-
tions, often tend in various ways, in the study of rocks
as well as in chemistry, to produce obscurity and not
unfrequently confusion. I was myself for some time

ON ITS EXTERIOR. VOLCANOES. 429

inclined to use the expressions orthoclase-trachytes,
labrador-trachytes, oligoclase-trachytes, and thus to
include glassy felspar (sanidine), on account of its che-
mical composition, in the genus orthoclase (common
felspar). The names were well-sounding and simple ;
but their simplicity was itself misleading; for, if the
term labrador-trachyte would correctly lead us to Etna
and Stromboli, that of oligoclase-trachyte, in its im-
portant twofold connection with augite and hornblende^
would falsely connect together the widely distributed,
very different formations of Chimborazo and the volcano
of Toluca. It is the association of a felspathic element
with one or two others, which here, as in the case of
certain dike formations, presents itself as the charac-
teristic indication.

The following classification of trachytes, distinguished
according to the crystals which are found enclosed in
them, has been made by Grustav Eose since the winter
of 1852. The principal results of this work, in which
oligoclase is never confounded with albite, were attained
by him 10 years before, when, in his geological examina-
tions in the Eiesengebirge, he found that oligoclase was
there an essential ingredient in the granite; and his
attention being thus directed to the importance of oligo-
clase as an essential constituent, he sought for it also in
other rocks. (594) This examination conducted him to
the important result that albite is never a constituent of
a particular kind of rock. (Poggend. Ann. Bd. Ixvi.
1845, S. 109.)

First Division. — "The ground-mass contains only
crystals of glassy felspar; these are in the form of
tablets, and are usually large. Hornblende and mica

430 BEACTION OF THE INTERIOR OF THE EARTH

are either entirely absent, or, at the utmost, occur only
sparingly, and are wholly unessential ingredients. To
this division belong the trachytes of the Phlegrsean
Fields (Monte Olibano, near Pozzuoli), of Ischia, and of
La Tolfa, and also of a portion of the Mont Dore
(Grande Cascade). Augite shows itself in small crys-
tals in trachytes of the Mont Dore, but only very
rarely (595); in the Phlegrsean fields, with hornblende,
not at all ; nor does leucite, of which, however, some
pieces were collected by Hoffmann above the Lago
Averno (on the road to Cumse), and some by myself on
the declivity of the Monte Nuovo(596), in the autumn of
1822. Leucite-ophyr, in loose pieces, is more plentiful
in the island of Procida and the neighbouring Scoglio
di S. Martino."

Second Division. — " The ground-mass contains de-
tached crystals of glassy felspar and a quantity of small
snow-white crystals of oligoclase. The latter are often
intermingled with the glassy felspar, and form a veil
round the felspar, as is often the case in Grustav Rose's
'granitite' (the rock which forms the main mass of
the Riesen- and Iser-gebirge, granite, with red felspar,
particularly rich in oligoclase and magnesia-mica ; but
without any white potassa mica). Hornblende and
mica, and, in some varieties, augite, sometimes present
themselves in small quantities. To this class belong
the trachytes of the Drachenfelg and Perlenhardt in the
Siebengebirge near Bonn (597), and many varieties of the
rocks of Mont Dore and Cantal : also several trachytes
in Asia Minor (of which we owe the knowledge to
Pierre von Tschichatscheff), at Afiun Karahissar (cele-
brated for the culture of the poppy), Mehammed Kjoe

ON ITS EXTEEIOK. VOLCANOES. 431

in Phrygia, and Kajadschyk and Donanlar in Mysia, in
which glassy felspar, much oligoclase, some hornblende,
and brown mica are intermixed."

Third Division. — "The ground-mass of this dioritic
trachyte contains many small crystals of oligoclase with
black hornblende and brown magnesia-mica. To this
division belong the trachytes of ^gina (598), of the
Kozelnik valley near Schemnitz (599), of Nagyag in
Transylvania, of Montabaur in the Duchy of Nassau, of
Stenzelberg and of the Wolkenburg in the Siebengebirge
near Bonn, of Puy de Chaumont near Clermont in
Auvergne, and of Liorant in the Cantal; Kasbegk in
the Caucasus, the Mexican volcanoes of Toluca (60°) and
Orizaba, the volcano of Purace, and, although it is
very uncertain whether these latter are trachytes, the
magnificent columns of Pisoje (601) near Popayan. Von
Buch's ( domites ' also belong to this division. In the
white fine-grained ground-mass of the trachytes of the
Puy de Dome there are glassy crystals which have
always been taken for felspar, but which are most dis-
tinctly streaked in the plane of cleavage, and are really
oligoclase; hornblende and some mica are found with
them. According to the volcanic specimens for which
the royal collection is indebted to Herr Mollhausen, the
draftsman and topographer of Lieutenant Whipple's
exploring expedition, we should also regard, as belonging
to this third division, or to the dioritic Toluca trachytes,
those of Mount Taylor, between Santa Fe del Nuevo
Mexico and Albuquerque, as well as those of Cieneguilla
on the western slope of the Rocky Mountains, where,
according to the fine observations of Jules Marcou,
black lava-streams are seen to have flowed over the

432 REACTION OF THE INTERIOR OF THE EARTH

Jurassic formation." The same mixtures of oligoclase
and hornblende which I have seen in the Aztec high-
lands (in Anahuac properly so called), but not in the
Cordilleras of South America, are also found far to the
westward of the Eocky Mountains and of Zurti, near
the Mohave Eiver, a tributary of the Eio Colorado. (See
Marcou, " Eesume of a Greological Eeconnaissance from
the Arkansas to California, July 1854," pp. 46 — 48 ; and
also two important French Memoirs entitled, " Eesume
explicatif d'une Carte Geologique des Etats-Unis, 1855,"
pp. 113 and 116, and "Esquisse d'une Classification des
Chaines de Montagues de 1'Amerique du Nord, 1855;"
"Sierra de San Francisco et Mount Taylor," p. 23.)
Among specimens from Java, which I owe to the kind-
ness of Dr. Junghuhn, we have also recognised trachytes
belonging to the " third division," from three volcanic
districts; those of Burung-agung, Tjin-as, and Grunung
Parang (in Batugangi).

Fourth Division.- — "The ground-mass contains augite
with oligoclase. To this division belong the Peak of
Teneriffe (602), the Mexican volcanoes Popocatepetl (603),
and Colima ; the South-American volcanoes Tolima
(with the Paramo de Euiz), Purace near Popayan, Pasto
and Cumbal (according to the fragments collected by
Boussingault), Eucu-Pichincha, Antisana, Cotopaxi,
Chimborazo (604), Tungurahua, and the trachyte-rocks
which are covered with the ruins of old Eiobamba. In
Tungurahua we also find, with the augites, some blackish
green uralite crystals, from half a line to five lines long
(from about one twentieth of an inch to half an inch),
with perfect augite-shape and the planes of cleavage of
hornblende. (See Eose, Eeise nach dem Ural, Bd. ii.

ON ITS EXTERIOK. VOLCANOES. 433

S. 353.) I brought such a piece with distinct uralite
crystals back with me from a height of 13,300 feet on
the side of Tunguragua. In Grustav Rose's opinion, it
is strikingly different from the seven trachyte fragments
from the same volcano which are in my collection, and
reminds him of the formation of schistose augitic por-
phyry which we found so extensively on the Asiatic
slope of the Ural. (See the last-quoted work, Bd. ii.
S. 544.)

Fifth Division. — «"A mixture of labradorite (605) and
augite (606) ; a doleritic trachyte. Etna, Stromboli, and,
according to the excellent examination of the trachytes
of the Antilles by Charles Sainte-Claire Deville, the
Soufriere de la Guadeloupe, and on Bourbon the three
great ( Cirques ' which surround the Pic de Salazu."

Sixth Division. — "An often-grey ( ground-mass,' in
which there are crystals of leucite and augite with very
little olivine : Vesuvius and Somma ; also the extinct
volcanoes of Vulturo, Eocca Monfina, the Alban Hills
and Borghetto. In the older masses, (for example in
the masonry and pavement of Pompeii,) the leucite
crystals are of considerable size and more abundant
than the augite. On the other hand in the present
lavas the augites prevail, and generally speaking leucites
are very rare. Nevertheless the stream of lava of April
22, 1845, furnished them in large quantities. (607) Frag-
ments of trachytes belonging to the first division,
containing glassy felspar (Leopold von Buch's ' trachytes
proper'), are found imbedded in the tufas of Monte
Somma ; and also singly beneath the layer of pumice
which covers Pompeii. The leucite-ophyr trachytes of
this sixth division are to be carefully distinguished from

VOL. IV. F P

434 REACTION OF THE INTERIOR OF THE EARTH

the trachytes of the first division, although, as has been
already noticed, leucites also occur in the westernmost
parts of the Phlegrsean Fields and on the island of
Procida."

The ingenious author of the classification of volcanoes
which has been here introduced, in which they are
classed according to the association of simple minerals
which they present, by no means considers that he has
exhausted the grouping of all that the surface of the
earth (still so exceedingly imperfectly explored in a
scientifically geological and chemical sense) can offer.
Alterations in the nomenclature of the associated mine-
rals, as well as an augmentation of our list of trachyte
formations, are to be expected in two ways : by the pro-
gressive improvement of mineralogy (in more accurate
specific distinction, at once according to form and ac-
cording to chemical composition), and by the augmenta-
tion of our collections, which are in most cases still so
incomplete and so little suited for the desired objects.
Here, as everywhere, when in cosmical considerations
the knowledge of the governing laws has to be attained
through a widely comprehensive comparison of particu-
lars, we must proceed from the fundamental principle
that all that we think we know in the present state of the
sciences is but a poor instalment of that which the next
century will present. A variety of means are now at
hand for accelerating the acquisition of such fuller
knowledge ; but there is still great deficiency, as regards
the application of thoroughly exhausting methods of
examination, in the exploration which has hitherto been
made of the trachytic portion of the upheaved, sunken,
or fissure-opened surface of the part of the earth which
is not covered by the ocean.

n

-

•

ON ITS EXTEBIOR. VOLCANOES. 435

Volcanoes which are very near to each other, and
which are similar in form, construction of framework,
and " geotechtonic " relations, have often a very different
individual character in the composition and association
of their mineral aggregates. On the great cross fissure
which, running almost due east and west from sea to
sea, intersects a south-east and north-west mountain
chain (or more truly an uninterrupted mountainous
swelling), — the volcanoes are arranged in succession
thus: Colima (13,003 feet), Jorullo (4265 feet), Toluca
(15,168 feet), Popocatepetl (17,726 feet), and Orizaba
(17,374 feet). The volcanoes which stand next to
each other are dissimilar in characteristic composition ;
the trachytes of the alternate ones are similar. Colima
and Popocatepetl consist of oligoclase with augite (there-
fore have the Chimborazo or Teneriffe trachyte) ; Toluca
and Orizaba consist of oligoclase with hornblende (there-
fore have the ^Egina and Kozelnik rock). The late
upheaved volcano of Jorullo, which can scarcely be
regarded as more than a great "hill of eruption,"
consists almost solely of basaltic and pitchy, mostly
scoriaceous lavas, and seems to be nearer to the Toluca
than to the Colima trachyte.

In these considerations on the diversity of minera-
logical constitution between neighbouring volcanoes,
we may at once see a censure of the unfortunate at-
tempt to introduce, as the name for a kind of trachyte,
a designation taken from a chain of mountains, in great
part volcanic, extending over a distance of 7200 geo-
graphical miles. The name of " Jura-limestone,"
which I was the first to introduce (G08), is not objection-
able, because it is taken from a simple, unmixed kind

FF 2

436 REACTION OF THE INTERIOR OF THE EARTH

of rock, and from a chain whose age is indicated by the
superposition of organic imbedded fossils ; nor is there
any objection to names taken from individual moun-
tains, or to the use of such expressions as " Teneriffe-"
<or " Etna-trachytes " for particular oligoclase or labra-
dorite formations. While there was a disposition to
recognise everywhere among the very different kinds of
felspar belonging to the trachytes of the Andes the
presence of albite, every rock in which it was supposed
to occur was termed " Andesite." I first find this name
with the positive definition, "Andesite is formed by
predominating albite and some hornblende," in an im-
portant memoir of Leopold von Buch, in 1835, "On
Craters of Elevation and Volcanoes." (609) This dis-
position to see albite everywhere lasted for five or six
years, until, in mpartially renewed and more thorough
examinations, the trachytic albites were recognised as
oligoclase. (61°) Grustav Eose has arrived at doubting
whether albite presents itself at all in these rocks as an
essential ingredient; and thus Andesite, according to
the older view of its character, would be wanting in the
chain of the Andes itself.

The mineralogical constitution of trachytes is im-
perfectly discerned when the porphyritically imbedded
crystals are not capable of being detached from the
ground-mass so as to be separately examined and mea-
sured, and when it is necessary to have recourse to the
numerical proportions of the earths, alkalies, and me-
tallic oxides, as given by analysis, and to the specific
gravity of the apparently amorphous mass which has to
be analysed. The desired result is obtained in a more
convincing and certain manner when both the ground-

ON ITS EXTERIOR, VOLCANOES. 437

mass itself and its contents can be separately examined,
both oryctognostically and chemically. This is the case,
for example, in the trachytes of the Peak of TenerifTe,
and in those of Etna. The assumption that the ground-
mass consists of the same minute undistinguishable
constituents which we recognise in the large crystals,
appears by no means solidly established ; for, as we have
already seen above, in Charles Deville's ingeniously exe-
cuted investigations, the apparently amorphous ground-
masses for the most part furnish more silicic acid than
would be expected from the felspar and other visible
ingredients. In leucite-ophyrs, as Grustav Eose remarks,
a striking contrast shows itself in the specific differences
of the prevailing alkalies, i. e. those of the inwoven
potassa-containing leucites, and of the ground-mass
which contains almost solely soda.(611)

But besides these associations of augite with oligo-
clase, augite with labradorite, and hornblende with
oligoclase, which have been instanced in the above
assumed classification of trachytes, and by which they
are specially characterised, we find in each volcano
other easily recognisable non-essential ingredients, whose
frequency or constant absence in different but often
neighbouring volcanoes is remarkable. It is probable
that this fact of the frequent occurrence of a particular
ingredient, or of its having only appeared at widely
separated intervals of time, has depended on a variety
of conditions, such as the depth at which the substances
have originated, temperature, pressure, degree of fluidity,
and slow or rapid cooling. The specific association or
absence of particular ingredients is opposed to certain
theories ; for example, to the supposed origin of pumice

438 EEACTION OF THE INTEEIOK OF THE EAHTII

from glassy felspar or from obsidian. These considera-
tions — which are by no means peculiar to very recent
times, having been touched upon towards the close of
the 18th century in reference to the comparison of the
trachytes of Hungary with those of Teneriffe — engaged
my earnest attention for several years, as appears from
my journals kept in Mexico and among the Andes.
Since that period great advances have been made in
lithology, and the less perfect determinations of mineral
species which I was then able to make have been either
corrected, or established on more secure foundations, by
Gustav Kose's long-continued examination and discus-
sion of my collections.

Mica.

Black or dark-green magnesia-mica is very frequent
in the trachytes of Cotopaxi, at a height of 14,470 feet,
between Suniguaicu and Quelendana, as well as in the
subterranean pumice-stone quarries of Guapulo and
Zumbalica, 16 miles from the foot of the volcano. (612)
The trachytes of the volcano of Toluca are also rich in
magnesia-mica, which is wanting at Chimb orazo. (613)
In our continent, micas have presented themselves in
abundance : at Vesuvius (for example, in the eruptions
of 1821 — 1823, according to Monticelli and Covelli) ;
in the Eifel in the ancient volcanic bombs of the
Lacher See(614); in the basalt of Meronitz, of the
marly Kausawer-Berg, and especially of the Gamayer-
Kuppe (615) in the Bohemian Mittelgebirge ; and, more
rarely in phonolite (616), as in the dolerite of the Kai-
serstuhl, near Freiburg. It is a remarkable circum-
stance, that in the trachytes and lavas of both continents

ON ITS EXTERIOR. VOLCANOES. 439

no white (mostly biaxial) potassa-mica, but only dark-
coloured (mostly uniaxial) magnesia-mica, appears to
have been produced ; and that this exclusive occurrence
of magnesia-mica extends to many other eruptive and
plutonic rocks, basalt, phonolite, syenite, syenite-slate,
and even to granitite ; while granite proper contains
simultaneously white potassa-mica and black or brown
magnesia-mica. (617)

Glassy Felspar.

This kind of felspar, which plays so important a part
in the activity of European volcanoes, in the trachytes
of the first and second divisions (for example, at Ischia,
in the Phlegrsean Fields, and in the Siebengebirge near
Bonn), is probably wholly wanting in the New Con-
tinent in the trachytes of active volcanoes ; which is the
more striking, inasmuch as sanidine (glassy felspar) be-
longs essentially to the argentiferous non-quartzose
Mexican porphyries of Moran, Pachuca, Villalpando,
and Acaguisotla, of which the first are connected with
the obsidians of Jacal. (618)

Hoi^nblende and Augite.

In the preceding account of the characteristics of the
six different divisions of trachytes, it was remarked that
the same mineral species which appear as essential ingre-
dients or constituents in some divisions (as, for example,
hornblende, in the third division, or Toluca rock), ap-
pear in other divisions (e.g. in the fourth and fifth,
the Pichincha and Etna rocks) only sporadically. I
have found hornblende, though not abundantly, in the
trachytes of the volcanoes of Cotopaxi, Euca-Pichincha,

440 REACTION OF THE INTERIOR OF THE EARTH

Tungurahua, and Antisana, by the side of augite and
oligoclase ; but on Chimborazo, up to a height of above
19,000 feet, I scarcely found any hornblende at all with
those two minerals. Of the many specimens brought
home from Chimborazo, hornblende has only been re-
cognised in two, and in those only in small quantity.
In the eruptions of Vesuvius in 1822 and 1850, augite
and hornblende crystals (the latter attaining a length of
between seven and eight tenths of an inch) were formed
contemporaneously by exhalations of vapours in fis-
sures. (619) On Etna, as Sartorius von Waltershausen
has remarked, hornblende belongs more particularly to
the older lavas. As the remarkable mineral which
Grustav Eose has called uralite, and which is widely
diffused in Western Asia and in several parts of Europe,
is nearly allied by structure and the form of its crystals
to hornblende and augite (62°), I would here notice that
uralite crystals have been recognised for the first time
as belonging to the New Continent, by having been
discovered by Eose in a piece of trachyte which I
knocked off on the declivity of Tungurahua, 3200 feet
below the summit.

Leudte.

Leucites — which in Europe belong exclusively to
Vesuvius, the Eocca Monfina, the Alban Hills near
Eome, the Kaiserstuhl in the Breisgau, and the Eifel
(on the western side of the Lacher See they are found
in blocks, not in the rock in situ, as on the Burgberg,
near Eieden) — have never been found in the volcanic
mountains of the New Continent and the Asiatic portion
of the old. Leopold von Buch had discovered, so long

ON ITS EXTERIOR. VOLCANOES. 441

ago as 1798, that they often form round a crystal of
augite, and had described this in an excellent me-
moir. (621) The augite crystal around which, according
to the remark of that great geologist, the leucite forms
is seldom wanting, but yet appears to be occasionally
replaced by a little nucleus or morsel of trachyte. The
unequal degree of fusibility between the kernel and
the surrounding leucite mass opposes some chemical
difficulties to the explanation of the mode of formation
within the envelope. Leucites, in part detached, ac-
cording to Scacchi, and in part intermixed with the lava,
were extremely abundant in the eruptions of Vesuvius
in 1822, 1828, 1832, 1845, and 1847.

Olivine.

Olivine is very frequent in the old lavas of Vesu-
vius (622), particularly in the leucite-ophyrs of the Somma ;
in the Arso of Ischia, in the eruption of 1301, mingled
with glassy felspar, brown mica, green augite, and mag-
netic iron ; in the Eifel volcanoes which have sent forth
lava-streams (e. g. in the Mosenberg, west of Mander-
scheid) (623) ; and in the south-east part of Teneriffe,
in the lava-eruption of Gruimar in 1704; but I have
sought for it zealously, but in vain, in the trachytes of
the volcanoes of Mexico, New Granada, and Quito.
Our Berlin collections contain from only four volcanoes,
— Tungurahua, Antisana, Chimborazo, and Pichincha, —
68 pieces of trachyte, 48 of which were brought home
by myself, and 20 by Boussingault.(624) In the basaltic
formations of the New Continent, olivine occurs as fre-
quently near augite as in Europe ; but the black basalt-

442 REACTION OF THE INTERIOR OF THE EARTH

like trachytes of Yana-Urcu, near Calpi, at the foot of
Chimborazo (625), as well as those belonging to what is
called " la reventazon del volcan de Ansango," (626) con-
tain no olivine. It was only in the great brownish-black
lava-stream, with a curled scoriaceous surface and cauli-
flower-like intumescence,, up which we made our way to
the crater of the volcano of Jorullo, that we found
small grains of olivine. (627) The very general scarcity
of this mineral in the more recent lavas and in the
greater part of the trachytes, becomes less surprising
when we remember that, however essential olivine may
appear to be, yet (according to Krug von Nidda and Sar-
torius von Waltershausen) there are in Iceland and in
the German Ehongebirge basalts without olivine, which
can scarcely be distinguished from the basalts having
olivine. In earlier times it was usual to call the former
"trap" and " wacke," and in more recent times, "ane-
masite." (628) Olivines, which are sometimes found as
large as a man's head in the basalts of Kentieres in
Auvergne, also attain, in the Unkel quarries, which
were the subject of my first youthful studies, a diameter
of more than 6 inches. The fine often-polished hyper-
sthene of Elfdal in Sweden, a granular mixture of
hypersthene and labradorite, which Berzelius has de-
scribed as syenite, also contains olivine (6-9), as does (a
still more rare case) the phono] ite of the Pic de
(jrriou (63°) in the Cantal. According to Strom eyer,
olivine is very constantly accompanied by nickel ; and
Kumler has discovered in it arsenic (631), a metal which
has very recently been found to be widely distributed
in mineral springs and even in sea-water. I have
spoken earlier of the occurrence of olivine in meteoric

1

ON ITS EXTEKIOE. VOLCANOES. 443

stones (aerolites) (632), and in artificial scorise (slags) (633),
examined by Sefstrom.

Obsidian.

So long ago as the spring and summer of 1799, when
I was preparing in Spain for my voyage to the Canaries,
the belief of the formation of pumice solely from ob-
sidian prevailed generally among the mineralogists of
Madrid, — Hergen, Don Jose Clavijo, and others. This
belief had been based on the study of fine geological
collections from the Peak of Teneriffe, as well as on the
comparison with phenomena in Hungary, although the
latter had then been for the most part represented
according to the interpretation resulting from the Nep-
tunian views of the Freiberg school. Doubts as to the
insufficiency of this theory of the formation of pumice,
which were very early awakened in me by my own
observations in the Canaries, in the Cordilleras of Quito,
and in the range of the Mexican volcanoes (634), incited
me to direct my most earnest attention to two groups
of facts: one being the general diversity of the sub-
stances enclosed in obsidians and in pumice-stones;
and the other their frequency of association or entire
separation in well-examined active volcanic frameworks.
My journals are filled with data on this subject ; and
my determinations of the species of the minerals form-
ing part of the rocks have since been assured by the
recent and varied examinations of my ever kind and
ready friend Gustav Eose.

In obsidian, as in pumice, glassy felspar is found as
well as oligoclase, and often both together. I may
adduce, as examples, the Mexican obsidians collected by

444 REACTION OF THE INTERIOR OF THE EARTH

me from the Cerro de las Navajas, on the eastern
declivity of Jacal ; those of Chico with many crystals of
mica ; those of Zimapan, S. S. W. of the city of Mexico,
with distinct, interspersed, small quartz crystals; the
pumices of the Rio Mayo (on the mountain-path from
Popayan to Pasto), and of the extinct volcano of Sorata
near Popayan. The subterranean quarries of pumice-
stone near Llactagunga (635) contain much mica, oligo-
clase, and (which is very rare in pumice-stones and
obsidian) also hornblende ; the latter has, however, also
been seen in the pumice of the volcano of Arequipa.
Common felspar, orthoclase, is never found in pumice
together with sanidine ; augites are also wanting. The
Somma, but not the cone of Vesuvius itself, contains
pumice which encloses earthy masses of carbonate of
lime. It is by the same remarkable variety of a calca-
reous pumice that Pompeii has been overwhelmed. (636)
Obsidians are rare in true lava-currents, having been
found almost exclusively in those of the Peak of Teneriffe,
Lipari, and Volcano.

Passing to the association of obsidian and pumice in
one and the same volcano, we find the following facts :
Pichincha has large fields of pumice, and no obsidian.
Chimborazo, like Etna — whose trachytes, however, have
quite a different composition (they contain labradorite
instead of oligoclase), — shows neither obsidian nor
pumice. I also remarked their absence in ascending
Tungurahua. The volcano of Purace, near Popayan, has
much obsidian intermixed in its trachytes, and has never
produced pumice. Large plains, as those from which Ili-
nissa, Carguairazo, and the "Altar " rise, are covered with
pumice. The subterranean pumice quarries near Llacta-

ON ITS EXTERIOE. VOLCANOES. 445

gimga, as well as those of Huichapa, south-east of Que-
retaro ; the pumice accumulations at the Rio Mayo (G37) ;
and those near Tschegem in the Caucasus (638) ; and near
Tollo (639) in Chili, at a distance from any active volcanic
frameworks, all appear to me to belong to the class of
phenomena of eruptions in the variously fissured flat
surface of the earth. Another Chilian volcano, that of
Antuco(640), of which Poppig has given a description
as scientifically important as it is pleasing in style,
produces, like Vesuvius, " ashes," small triturated rapilli
(sand), but no pumice, and no vitrified or obsidian-like
rock. Without the presence of obsidian or of glassy
felspar, and with great diversity of composition in the
trachytes, we see both the production and the non-
' production of pumice. Pumice, as the sagacious Darwin
has observed, is, moreover, wanting in the entire group
of the Galapagos. We have already remarked elsewhere
that in the great volcano of the Sandwich Islands, Mauna
Loa, as well as in the volcanoes of the Eifel, which once
poured forth lava(641), cones of ashes are wanting.
Although the island of Java has a range of more than
forty volcanoes, of which twenty-three are now active,
yet Junghuhn has only been able to discover two points
— at the volcano Grunung Gruntur, not far from Bandong,
and in the great Tengger mountains (642) — where masses
of obsidian have formed. They do not appear to have
occasioned any formation of pumice. The seas of sand
(dasar), which are at a mean height of nearly 7000 feet
above the sea, are covered, not with pumice, but with a
layer of rapilli, which are described as half-vitrified
obsidian-like pieces of basalt. The cone of Vesuvius,
which has never emitted pumice, sent forth, from the

446 REACTION OF THE INTERIOR OF THE EARTH

24th to the 28th of October 1822, sand-like ashes,
triturated trachytic rapilli, forming a layer or stratum
19 inches thick, which have never been confounded
with pumice.

The hollows and vesicular cavities in obsidians in
which (as, for example, at the Mexican Cerro del Jacal)
crystals of olivine have formed, probably from precipi-
tated vapours, are occasionally found in both hemi-
spheres to contain another kind of enclosures, which
seem to point to the mode of their origin and forma-
tion. In the wider portions of these long-extended and
mostly very regularly parallel cavities, there are morsels
of half-decomposed earthy trachyte. The cavity has a
narrower tail-like prolongation, as if, under the influ-
ence of volcanic heat, a gaseous elastic fluid had deve-
loped itself in the still soft mass. This phenomenon
had particularly attracted the attention of Leopold von
Buch in 1805, when he, Gray-Lussac, and I visited
Thomson's collection of minerals at Naples. (643) The
puffing up of obsidians by fire, which had not escaped
observation in Grecian antiquity (644), is certainly caused
by a similar development of gas. According to Abich,
obsidians submitted to fusion pass so much the more
easily into cellular pumice, in which the fibres are not
parallel to each other, the poorer they are in silicic acid
and the richer they are in alkalies. But whether the
inflation is due solely to the flying off of potassa or
hydrochloric acid remains, according to Eammelberg's
investigations (645), very uncertain. Apparently similar
phenomena of inflation in trachytes rich in obsidian and
sanidine, in porous basalts and amygdaloids, in pitch-
stone, tourmaline, and the dark-brown flint which

ON ITS EXTERIOR. VOLCANOES. 447

loses colour, may have very different causes in dif-
ferent substances ; and an examination long looked for
in vain, based on appropriate and exact experiments
made exclusively on the escaping gases, would lead to
an inestimable enlargement of the chemical geology of
volcanoes, if regard were at the same time paid to the
operation of sea-water in submarine formations, and to
the quantity of carburetted hydrogen in the intermixed
organic substances.

The facts which I have brought together at the close
of this section — the enumeration of volcanoes which pro-
duce pumice without obsidian, or much obsidian without
pumice, and the remarkable, not constant, but, on the
contrary, very various association of obsidian and pumice
with certain other minerals — led me, long ago, during
my sojourn in the Cordilleras of Quito, to the persua-
sion that the formation of pumice is the result of a che-
mical process, which may take place in trachytes of very
various composition, without the necessary intervention
of obsidian (i.e. without the necessary preceding existence
of obsidian in large masses). The conditions under
which such a process goes forward on a great scale are,
probably, — I would here repeat, — dependent less on
diversity of materials, than on degrees of heat, of pres-
sure determined by depth, of fluidity, and on length of
period of solidification. The highly noteworthy although
rare phsenomena presented by the isolated occurrence of
vast subterranean pumice-stone quarries, apart from any
" volcanic framework," lead me to entertain the conjec-
ture^46) that a not inconsiderable portion — perhaps,
according to volume, the larger portion — of volcanic
rocks have been emitted, not from elevated volcanic

448 EEACTION OF THE INTERIOR OF THE EARTH.

frameworks, but from a network of fissures on the
earth's surface, from which they have poured forth,
often forming, as it were, strata covering an extent of
many square leagues. We are probably to include
among these phenomena the old masses of trap of the
lower Silurian formation of the south-west of England,
by whose exact chronometric determination my illus-
trious friend Sir Roderick Murchison has so greatly
enlarged the scope and heightened the character of our
knowledge of the geological construction of the globe,

449

RECTIFICATIONS AND ADDITIONS.

Page 32.

A STILL much higher result for the earth's density than
was obtained by Baily (1842) and by Reich (1847-1850)
has been given by the pendulum experiments which
Airy instituted, with exemplary precautions, in 1854,
in the Harton coal-mine. The result of these experi-
ments is to give the density 6*566, with a probable
error of 0-182 (Airy, in the Phil. Trans. 1856, p. 342).
A small modification of this numerical value is added
by Professor Stokes, on account of the effect of the
earth's rotation and ellipticity, and alters the density to
6-565 for Harton, which is in 54° 48' K lat., and to
6-489 for the equator.

Page 80, line 5.

Arago left behind him a rich mass of magnetic obser-
vations (more than 52,600 in number), obtained from
1818 to 1835, which, after a laborious redaction by
M. Fedor Thoman, have been published in the "GEuvres
completes de Francois Arago" (tome iv. p. 498).
General Sabine has discovered in -these observations (for

VOL. IV. G G

450 RECTIFICATIONS AND ADDITIONS.

the years 1821 — 1830) the most entire confirmation of
the decennial magnetic declination period, and its con-
nection with the similar period of relative paucity and
frequency in the solar spots. (Arago's Meteorological
Essays (Translation), London, 1855, pp. 355 — 357.) As
early as the same year (1850) in which Schwabe, at
Dessau, published his periods of the solar spots (Kosmos,
Bd. iii. S. 402 ; Eng. p. 291), and even two years before
Sabine himself first (in March 1852, Phil. Trans, for 1852,
pt. 1, p. 116 — 121 ; Kosmos, Bd. iv. S. 174 ; Eng. note
73) connected the magnetic decennial period with that
of the solar spots, he had made the important discovery
that the sun appears to act by its own proper magnetic
force upon the earth's magnetism. He had found (Phil.
Trans. 1850, pt. 1, p. 216 ; Kosmos, Bd. iv. S. 132 ; Eng.
p. 143) that the magnetic intensity is greatest, and the
needle approaches most nearly to a vertical direction,
when the earth is nearest to the sun. The knowledge
of such a magnetic influence of the central body of oar
planetary system, not acting indirectly through the pro-
duction of heat, but directly by its own magnetic force,
as well as by variations in its photosphere (in the mag-
nitude and frequency of funnel-shaped openings in the
solar photosphere), gives to the study of terrestrial
magnetism, and to the network of magnetic observa-
tories with which (Kosmos, Bd. i. S. 346, Bd. iv. S. 72 ;
Eng. vol. i. p. 334, present volume, p. 78) Russia and
Northern Asia have been covered since the resolutions
of 1829, and the colonies of Great Britain since
1840 — 1850, a higher cosmical interest. (Sabine, in
the Proceedings of the Royal Society, vol. viii. No. 25,
p. 400, and in the Phil. Trans, for 1856, p. 362.)

RECTIFICATIONS AND ADDITIONS. 451

Page 90, line 4.

Although the nearness of the moon in comparison
with the sun does not seem to compensate the smallness
of its mass, yet the already securely ascertained variation
of the magnetic declination in the course of a lunar day,
"lunar-diurnal magnetic variation" (Sabine in the
Report to the British Association at Liverpool, 1854,
p. 11 ; and for Hobarton, Phil. Trans. 1857, art. i. p. 6),
incites us perseveringly to continue and extend the
researches on the magnetic influences of our satellite.
Kreil has the great merit of having pursued this ex-
amination with great care from 1839 to 1852 (see his
"Abhandlung iiber den Einfluss des Mondes auf die
horizontale Componente der magnetischen Erdkraft," in
the " Denkschriften der Wiener Akademie der Wiss.
math em. naturwiss. Classe, Bd. v. 1853, S. 45 ; and Phil.
Trans, for 1856, art. xxii.) As M. Kreil's observations,
made during several years at Milan and Prague, led him
to believe that the moon causes a decennial declination
period similar to that occasioned apparently by the sun-
spots, General Sabine was induced to undertake a more
extensive investigation, whereby he ascertained that the
influence of the sun alone produces a decennial period,
which indeed he had already made out for Toronto, in
Canada, by the employment of a special and very accu-
rate mode of calculation (Phil. Trans. 1856, p. 361).
This influence of the sun was also distinctly manifested
by the hourly observations, continued during eight years,
at Hobarton in Van Diemen Island ; but while, in both
hemispheres, the southern as well as the northern, the

GO 2

452 RECTIFICATIONS AND ADDITIONS.

same result was found for the sun's action, they also both
agreed in yielding assured evidence " that the lunar-diur-
nal variation corresponding to different years shows no
conformity to the decennial inequality manifested in
those of the solar-diurnal variation." No sufficient ex-
planation seems yet to have been given of the lunar-
diurnal magnetic variation, as General Sabine has also
remarked that "the earth's inductive action reflected
from the moon would seem to be too feeble to account
for it." (Sabine in the Phil. Trans, for 1857, art. i. p. 7 ;
and in the Proceedings of the Koyal Society, vol. viii.
No. 20, p. 404.) As the magnetic portion of this volume
was printed almost three years ago, it appeared to be
particularly needful, in regard to a subject to which I
have been so long attached, to complete it, in some
measure, by some subsequent additions.

453

EDITOR'S NOTES.

I. — On the Ellipticity of the Earth.

IN the 28th and 29th pages of the German original of this
volume (English translation, pp. 27 and 28), M. de Hum-
boldt has combined the results of the pendulum experi-
ments made by M. Biot, in conjunction with MM. Arago,
Chaix, and Mathieu, at several stations of the French
arc of the meridian, with those made by myself at several
stations in Europe, Africa, and America, (chosen chiefly
with reference to their proximity either to the equator
on the one hand, or to the high latitudes of the northern
hemisphere on the other,) for the purpose of drawing a
conclusion upon the important question, whether the
acceleration of the pendulum between the equator and the
middle latitudes, and between the middle latitudes and
the vicinity of the pole, indicates a uniformity, or an
inequality, in the amount of the ellipticity derived from
those two portions of the hemisphere. The conclusion
arrived at is, that from the equator to the latitude of 45°
the pendulum results indicate T^i and from latitude
45° to the pole T J-g- ; whilst the three series taken toge-
ther give for the whole northern quadrant the ellipticity

454 EDITOR'S NOTES.

But before such a combination as is here made can
be safely employed in the deduction of the ellipticity
of the whole quadrant, and still more of separate por-
tions of the quadrant, it is of the first importance to
be assured of the strict intercomparability of results
obtained in the different parts of the hemisphere by dif-
ferent methods and by very dissimilar apparatus.

The pendulum experiments of the French philosophers
in the middle latitudes were made, as is well known, by
the method and with the apparatus (slightly modified)
of Borda. Mine, both in the neighbourhood of the
equator and in the high latitudes, with the very different
apparatus known as the Invariable Pendulum of Kater.
The object which is sought by the experiment is also
different in the two cases. With the apparatus of Borda,
the object sought, and the result obtained, is the absolute
length of the seconds pendulum at the place at which
the experiment is made. The object sought, and the
result obtained, by the invariable pendulum is a far
more simple one, viz. the acceleration of the pendulum
in the different latitudes to which the pendulum is suc-
cessively conveyed. The acceleration is a function of
the ellipticity ; and the ellipticity is deduced from the
acceleration, quite independently of the absolute length
of the seconds pendulum at any one of the stations
of experiment ; which length is neither determined
nor determinable by means of the invariable pendu-
lum, nor does it need to have been determined at any
station whatsoever. To bring, consequently, a series
of experiments with an invariable pendulum (which
furnish simply the acceleration, or the difference in
the number of vibrations which a pendulum media-

ON THE ELLIPTICITT OF THE EARTH. 455

nically invariable makes in a certain definite time at
the different stations to which it has been successively
conveyed,) into combination with another series in which
the experimental object has been a different one, — viz.
to determine, by Borda's method, the absolute length of
the seconds pendulum at one or more stations, — requires
that some assumption should be made, whereby the
absolute lengths of the seconds pendulum at the stations
where the invariable pendulum has been used may be
substituted for that which is the sole and simple ex-
perimental result with the latter apparatus ; and,
further, that the lengths so substituted shall be such as
would have been obtained by Borda's apparatus in
particular, had it been employed at those stations in-
stead of the invariable pendulum.

The operation of even a very small error in the
requisite assumption may easily be understood when,
as in the present case, the one series — viz. that of
the invariable pendulum — has furnished the results
in both the polar and the equatorial latitudes, and the
other series — viz. that of the absolute lengths measured
by Borda's apparatus — those in the middle latitudes.
1st, By the combination of the polar and equatorial
results we obtain the ellipticity unaffected by the sup-
posed error, because they form parts of one and the
same experimental series, and require no substitution
in lieu of the direct experimental results. They have
also the very great and independent advantage of com-
prehending between them an arc of such extent (and
consequently an acceleration of such amount), that
minor errors of other kinds, whether observational, or
due to local variations of the gravitating force as ill-

456

fluenced by the disposition or density of the materials at
the surface, become of comparatively minor importance.
The ellipticity derived from this combination is there-
fore not only unaffected by the supposed error of which
we are treating, but is also much less liable to be
affected by errors of other kinds, than are the results of
the two following combinations in which the arcs are of
inferior extent. 2nd, In combining the middle latitude
series with one portion — say, the equatorial — of the
other series, the error of the assumption, if there be
one, is involved ; and though it may be a very small
error, its influence on the ellipticity derived from this
combination may be comparatively large, because the
arc comprehended by the stations is now reduced to
nearly one half of the extent it had when the polar and
equatorial series were taken together. 3rd, The same
reasoning holds when the middle latitude series is com-
bined with the other portion, i. e. the polar portion, of
the series with the invariable pendulum: but in this
case the ellipticity resulting from the combination will
have an error in the opposite sense to the deduction in
No. 2. 4th and lastly, When the whole of the results
of the two series are combined, by means of an assump-
tion which involves a small error in the comparability of
the middle latitude results with those of the polar and
equatorial stations, the ellipticity resulting therefrom will
exhibit a small difference from that derived from the
combination of the polar and equatorial results, for the
strict comparability of which no such assumption was
required ; but the difference will be small in comparison
with that in 2 and 3, partly because opposite effects
tend to counterbalance each other, and partly because

ON THE ELLIPTICITT OF THE EAETH. 457

all differences are softened down by the extent of the
comprehended arc.

Now the ellipticities derived from the four combina-
tions are as follows: — 1. Polar and equatorial stations,
__L._; (Sabine, Pend. Ex. 1825, p. 334.) 2. Equa-
torial and middle latitudes, -^-g ; (Kosmos, B. iv. S. 28 ;
Eng. tr. vol. iv. p. 28.) 3. Polar and middle latitudes,
-gi-g-; (Kosmos, idem.) 4. Polar and equatorial with
middle latitude stations, -§fa ; (Kosmos, idem.) Com-
paring the differences thus shown with the statement
that has preceded them, we may perceive that they
correspond precisely to what would be occasioned by an
error in some part of the process, by which a strict in-
tercomparability has been supposed to be established,
between the results obtained by myself with the invari-
able pendulum in the equatorial and polar latitudes, and
those obtained in the middle latitudes with Borda's
apparatus.

From a conviction of the great difficulty of establish-
ing an unexceptionable intercomparability between the
results of experiments in which the objects, methods,
and apparatus were dissimilar, I did not hesitate, when
investigating in 1825 the same question of uniformity
or otherwise of the ellipticity in the two portions of the
northern quadrant, to. prefer the combination of Captain
Kater's results at the principal stations of the British arc
of the meridian with my own in the equatorial and high
northern latitudes, to a combination of the experiments
on the French arc with my own. In the first combination,
that of Captain Kater's experiments and mine, the appara-
tus was the same; the method of experimenting identical ;
the connection between our separate series such as could

458

•

not possibly be better — we had the same base station
(London), the same house in London, and the same
apartment in the same house : the results at our several
stations formed, in fact, one series, of unquestionable
relation in every particular. For the solution of the
special question which we are now considering they fur-
nished me with three groups, remarkably well adapted
for the purpose ; the first comprising the five stations
nearest to the equator, i. e. between 0° and 10° 38' 50"
(Trinidad); the second comprising the six stations in
England and Scotland ; and the third group comprising
the five most northern stations, i. e., between 60° N.
and 80° N. The ellipticity severally derived from the
combinations of these three groups (which appears to
have escaped the notice of M. de Humboldt) was as
follows : —

Equatorial and middle latitude - ^TS-T 1 Sabine,
Equatorial and high northern - -j-^.-^ > Pend. Exp.
High northern and middle latitude jt^.yJ 1825, p. 346.

It is seen, therefore, that when the experiments
are made under suitable and favourable conditions of
Uniformity in the apparatus and method, and of well
assured relation to each other throughout, the supposed
" anomaly" in the ellipticity of the two portions of the
quadrant entirely disappears ; each portion of the quad-
rant corresponds with the other in the amount of
ellipticity which it indicates, and both correspond with
the deduction made for the whole quadrant from nine-
teen stations distributed over its surface, extending
from the equator to a few minutes short of 80° of K
latitude.

ON THE ELLIPTICITY OF THE EARTH. 459

When the object of experiment is simply the de-
termination of the ellipticity, the invariable pendulum
unquestionably presents the most direct and the
most safe method of ascertaining the acceleration in
different latitudes, from which the ellipticity is an im-
mediate deduction. It is far preferable to an inde-
pendent determination, at each station of experiment,
of the absolute length of the seconds pendulum, either
by Borda's method, or by Kater's convertible pen-
dulum; and still more to an endeavour to combine
results at different stations which have been arrived at
by the employment of dissimilar apparatus and pro-
cesses. The determination of the absolute length de-
mands a far more elaborate and delicate process of
experiment than is required simply for the acceleration ;
and, since Bessel's remarkable discovery in 1828
(Kosmos, Bd. iv. Anm. 20), it has been known that one
part of the process, whereby the absolute lengths were
supposed to be determined before that date, was founded
on an erroneous theoretical supposition ; and hence has
followed the necessity, wholly unforeseen and unsus-
pected by the eminent persons who devised the methods
or who employed them previous to 1828, of experi-
mentally investigating and applying corrections by no
means insignificant in amount, and affecting different
forms of apparatus in very different proportions.

Although it is obvious on a very slight consideration,
that the corrections required by Bessel's discovery to be
applied to the acceleration, derived from experiments
with an invariable pendulum in which the same form of
pendulum had been employed throughout, must be very
small in comparison with those required for the absolute

460 EDITOR'S NOTES.

lengths referred to in the preceding paragraph, yet, as
the conclusions which have been here reproduced are
taken from my publication in 1825 (which publication
was anterior to 1828, the epoch of the announcement of
Bessel's discovery), it may be proper on this occasion
to state explicitly the degree in which those conclu-
sions were affected by it. It is conformable to the
plan of this volume of e ( Kosmos " (which treats of the
" special results of observation in the domain of the ter-
restrial phenomena "), that it should contain the data
upon which important conclusions are based, corrected
according to the most recent emendations. Now Bessel
has shown that when a pendulum moves in air the air
becomes part of the moving system, and the moving
force must be imparted, not only to the particles of
mass of the solid body of the pendulum, but also to
all the particles of mass of the air which are put in
motion in accompaniment with it. The specific gravity
of the moving mass, including solid and gaseous, being
less than that of the solid part alone, the method em-
ployed antecedently for computing the effect of the me-
dium in which the pendulum was vibrated, in retarding
the vibration, was erroneous ; and, as the quantity and
distribution of the air accompanying the solid- portion of
the pendulum must depend upon the form and ar-
rangement of the solid part, it followed that the true
"correction for buoyancy" for each particular form of
experimental pendulum required to be ascertained by
special experiment. It was obvious that in the case of
the determinations of absolute length, as the apparatus
of Borda and that of Rater's convertible pendulum
differed widely in form, their respective corrections for

ON THE ELLIPTICITT OF THE EARTH. 461

buoyancy might be expected to differ widely from each
other, as well as from the original, and now known to be
erroneous, computation ; and that the results would, with
each apparatus, be altered by the whole amount of the
difference between the original computation, and the true
buoyancy correction as learnt by experiment ; whilst in
the case of the invariable pendulum the acceleration
would only require to be corrected for a small fraction
of that difference, because the greater portion would be
a common error at all the stations where the same form
of invariable pendulum had been used.

The kind of experiment best suited to give the true
"correction for buoyancy" (or it is sometimes called
" the reduction to a vacuum ") for any particular form
of pendulum, was sufficiently obvious, and presented no
other difficulties than such as might be surmounted by
a fitting apparatus, in which the pendulum itself might
be vibrated alternately in air of the ordinary tempera-
ture and density, and in rarefied air approaching nearly
to a vacuum.

In the same year that Bessel's discovery was an-
nounced (1828), I had such an apparatus constructed,
and, by experiments recorded in the Phil. Trans, for
1829, art. xviii., ascertained that the buoyancy correc-
tion for an invariable pendulum of the form employed
by Captain Kater and myself, computed by the formula
which had been previously in use, required to be multi-
plied by the factor 1-65 in order to give the true reduction
to a vacuum. The most widely differing buoyancy cor-
rections at the 19 stations of Captain Kater and myself,
Computed by the original formula, were: + 5S*75 at
Sierra Leone, and + 6S*27 at Spitzbergen, in a mean

462 EDITOR'S NOTES.

solar day. These corrections, multiplied by 1*65, be-
come respectively 9S*52 and 10S*38 ; or the number of
vibrations in a mean solar day, computed from the
results with the invariable pendulum before Bessel's
discovery was made, required to be increased by
(9s-52 - 58-7o =) 3s-77 at Sierra Leone, and (108-38

— 68'27 = )4S<11 at Spitzbergen. But the acceleration
between these stations would only have to be corrected
for the difference between these numbers, i.e. for (4s* 11

— 3s-77=) 08*34 ; and as the whole acceleration be-
tween Sierra Leone and Spitzbergen is 214S>8 in a mean
solar day, an error of Os>34, or about its -g-J-oth part, would
be of very small import. But it so happens that even
this small error is in great part compensated by a nearly
equivalent correction in an opposite sense, which is oc-
casioned in the reduction to a uniform temperature as a
consequence also of Bessel's discovery. The equivalent
in seconds to 1° of Fahr. was inferred to be 08*421 in a
mean solar day, from experiments made in London
previously to 1828, in temperatures respectively 45°-81
and 83°-55 of Fahr. (Pend. Expts., pp. 202—207.) In
this deduction the old and erroneous buoyancy formula
was employed; but, on a recomputation with the old
correction multiplied by 1*65, the true equivalent for
1° is found to be 08-43. (Sabine, in Phil. Trans., 1829,
p. 238 ; and Airy, in Phil. Trans., 1856, p. 317.) Now
the temperature of the pendulum during the experi-
ments at Sierra Leone was 80°%52, and at Spitzbergen
41°*13, the difference being 39°*39, which being multi-
plied by (08-43-08-421 =) OH)09 is 08-36 ; by which
amount the originally computed acceleration between

ON THE ELLIPTICITY OF THE EAKTH. 463

Sierra Leone and Spitzbergen should be diminished on
account of the more accurate correction for the differ-
ence in the temperature of the pendulums, as it was
previously shown that it ought to be increased by
Os'34 as a more accurate correction for the reduction to
a vacuum ; both the more recent values of the correc-
tions being consequences of Bessel's discovery. Sierra
Leone and Spitzbergen are stations of nearly extreme
difference, both in the temperature of the pendulums
and in the temperature and density of the air: the
effect of Bessel's discovery on the acceleration as ori-
ginally computed, and, consequently, on the ellipticity
originally deduced, is seen, therefore, to be wholly
insignificant.

Thus it has happened that, — from my having adopted
what appeared to me a more practical mode of ex-
amining the influence of differences of temperature on
the time of vibration of the pendulums, viz., by caus-
ing them to vibrate at one and the same station in
temperatures differing so widely as to include the whole
range of temperature experienced elsewhere, — (instead
of deriving the correction from the expansion of brass
measured by pyrometric experiments, as previously
practised), — the acceleration originally computed from
my experiments in different latitudes has stood in need
of little (and that little quite unimportant) correction as
a consequence of Bessel's discovery in 1828. The reduc-
tions to a uniform temperature are indeed everywhere too
small, if regarded as representing only the effect of the
expansion of the metal on the vibrations of the pendu-
lum ; but they are experimentally all but correct when

464

regarded as representing the joint effects of the expan-
sion of the metal, and of the difference of temperature
on that part of the reduction to a vacuum which is not
comprehended in the ancient " correction for buoyancy."
(Phil. Trans., 1829, p. 238.)

Although the correctness or incorrectness of the mea-
surements of the absolute length of the pendulum,
which were made before 1828, has no bearing on the
ellipticity derived, as here, from direct experiments on
the acceleration, it may be well to notice the great com-
parative influence which Bessel's discovery has on one
at least of the methods, by which the true length of the
seconds pendulum was previously supposed to have been
determined ; whence we may at once perceive the inse-
curity of inferring supposed comparable values of the
acceleration from absolute lengths measured in different
places by different methods. Experiments made in 1830,
and reported in the Phil. Trans, for 1831, art. xxv., with
the vacuum apparatus already spoken of, and with the
identical convertible pendulum which had been used by
Captain Kater (in his experiments on the length of the
seconds pendulum in London in 1817), showed that
the true reduction to a vacuum of that pendulum
was more than double the amount calculated by the
formula since known to be erroneous. The true reduc-
tion was between 12 and 13 seconds in the 24 hours,
instead of about 6 seconds, as computed and employed
by Captain Kater; the error in his determination was
consequently more than 6 seconds in a day, equivalent
to '006 parts of an inch in the length of the seconds
pendulum. In the case of the convertible pendulum,

ON THE ELLIPTICITY OF THE EAETH. 465

the amount of the true reduction to a vacuum was no
doubt unusually large in consequence of its particular
form, and therefore it can be no guide in regard to the
similar reduction required in other experimental forms,
which must in every case be ascertained by special ex-
periment. This is one of many points which require
to be investigated, before measurements of the absolute
length of the seconds pendulum, by different methods
and dissimilar apparatus, can be assumed with confi-
dence as giving identical conclusions.

The variations in the rate of the pendulum at the
stations of Captain Kater and myself are shown in the
following tabular view. They are expressed in refer-
ence to the rate in London (86,400 seconds in a mean
solar day), in the house which was the common base
station of both experimentalists. The pendulum being
by its mechanical construction invariable in its length
(excepting only from the effects of changes of tempera-
ture), the variations in its rate are the direct results of
observation at the several stations, requiring only three
corrections to be applied : — 1. for differences of tem-
perature ; 2. for differences of the density of the air ;
3. for differences of elevation above the level of the sea.
For the first and second, the corrections have themselves
been determined experimentally ; and for the third, the
correction has been calculated from the duplicate pro-
portion of the distances of the station and of the sea-
level from the earth's centre, diminished by a uniform
factor of 0-67. (Phil. Trans., 1819, p. 354; and Pend.
Exp., p. 332.) Of the three corrections, that for the
variations of temperature is the one of which the ac-

VOL. iv. HH

466 EDITOR'S NOTES.

curacy may be the most open to question. The value
Os'43, as the equivalent of a change of 1° of Fahr. in
the temperature of the pendulum, was obtained in
the experiments of 1824, by the comparison of vibra-
tions in temperatures which differed from each other
nearly the whole amount of the range of the natural
temperatures experienced at the several stations, and
measured by the same thermometers. But as those
experiments were made in the middle of a London
winter, the high temperatures were necessarily produced
by heating the apartment by artificial means ; and al-
though this was done with great care, and in many
respects under very favourable circumstances, it is
always difficult to maintain for many hours an artificial
temperature, exceeding that of the atmosphere by little
less than 40°, so as to be quite sure of a correct mean
temperature. For this reason I was glad to have an
opportunity of examining the question afresh in 1829 —
1830, by experiments made at the Royal Observatory at
Greenwich (with the permission of Mr. Pond, then astro-
nomer royal,) with a pendulum similar to those I had
previously employed, and with the same thermometers,
in natural temperatures varying about 32° of Fahr. from
each other. These are recorded in the Phil. Trans, for
1830, art. xix. They gave the value corresponding to
1° of Fahr. 08-44 instead of Os-43, the latter having
been obtained from the comparison of the vibrations in
the natural and artificial temperatures. In submitting
the experiments of 1829—1830 to the Royal Society, I
expressed my own preference for the value, Os-44, ob-
tained when both the extremes were natural tempera-

ON THE ELLIPTICITY OF THE EARTH. 467

tures. In restating the general results, therefore, on
the present occasion, I have employed Os'44 as the
equivalent of 1° Fahr. in reducing both my own and
Captain Kater's results to a uniform temperature. The
" reduction to a vacuum " is the ordinary buoyancy cor-
rection multiplied by the factor 1'65 ; and in the reduc-
tion to the sea-level the factor 0-67 has been applied
uniformly at all the stations ; the heights were gene-
rally small, and the reductions little more than nominal.
Whatever errors in the original publication have come
to my knowledge subsequently, either by my own recal-
culations or by information received from others, have
been corrected ; and three stations have been added to
the original nineteen, viz. the observatories of Green-
wich, Altona, and Paris, which, at the request of the
council of the Eoyal Society, I visited subsequently to
the polar and equatorial series, employing pendulums of
precisely the same form and construction, and pursuing
the same processes both of observation and reduction.
(Phil. Trans., 1828, art. iv.; 1829, art. ix. ; 1830,
art. xviii.) The reductions to the sea-level have been
applied to the results in the three memoirs just cited,
which reductions were not required in the first instance,
because the object then sought was the difference in
the rate of vibration at the three observatories from the
rate in London. The same coefficients have been em-
ployed in the corrections for temperature and buoyancy
as at the other stations. On all occasions the inva-
riability of the pendulums during the experiments was
proved, by bringing them back to the base station, and
ascertaining that their rates were unchanged.

HH 2

468

EDITOR'S NOTES.

VARIATIONS IN THE RATE OF THE PENDULUM VIBRATING
SECONDS or MEAN SOLAR TIME IN LONDON WHEN TAKEN
TO OTHER LATITUDES : —

Latitude.

Rate.

Observer.

St.. Thomas . . .
Maranham . . .
Ascension . . .
Sierra Leone . .
Trinidad . . .
Bahia ....
Jamaica . . . .
New York . . .
Paris ....

0 24 41 N.
2 31 34 S.
7 55 30 S.
8 29 28 N.
10 38 55 N.
12 59 21 S.
17 56 7 N.
40 42 43 N.
48 50 14 N.

losing 130-68 Sabine.
„ 140-23 Sabine.
„ 127-96 Sabine.
„ 131-67 Sabine.
„ 132-73 Sabine.
„ 126-84 Sabine.
„ 114-88 Sabine.
„ 42-27 Sabine.
„ 11*48 Sabine.

Shanklin . . .
Greenwich . . .

London . . . .

Arbury ....
Clifton ....
Altona ....
Leith

50 37 24 N.
51 28 40 N.

51 31 8 N".

52 12 55 N.
53 27 43 N.
53 32 45 N.

55 58 41 N.

„ 3-46 Kater.
gaming 0'59 Sabine.
f Kater.
» °'° {Sabine.
„ 3*31 Kater.
„ 7-23 Kater.
„ 8-94 Sabine.
„ 17-89 Kater.

Portsoy ....
Unst . . . . ..

57 40 59 N.
60 45 28 N.

„ 24-60 Kater.
„ 35-52 Kater.

Drontheim . . .
Hammerfest . ..
Greenland . . ..
Spitzbergen . .

63 25 54 N".
70 40 5 N.
74 32 19 N.
79 49 54 N.

„ 38-77 Sabine.
„ 61-05 Sabine.
„ 70-50 Sabine.
„ 83-01 Sabine.

Proceeding by well-known methods, we obtain, from
the data contained in the preceding table, 86263-60 as
the corresponding rate of vibration at the equator, and
•0051828 as the increase of gravity from the equator to
the pole; together with the ellipticity, ^Vl> correspond-
ing precisely with the value derived from the thirteen
stations of my equatorial and arctic voyages. (Pend.
Expts., 1825, p. 334.)

The following table exhibits the vibrations at the

ON THE ELLIPTICITY OF THE EARTH.

469

several stations computed by and corresponding to these
values of the equatorial rate of vibration, and of the
increase of gravity from the equator to the pole ;
together with the vibrations actually observed at each
station, and the differences, or what may be properly
termed the discrepancies, of the observed vibrations :—

Vibrations.

Differences.

Stations.

Computed.

Observed.

Equator . . . .
St. Thomas . . .
Maranham . . .
Ascension . . .
Sierra Leone . .
Trinidad ....
Bahia

86263-60
86263-60
86264-30
86267-86
86268-48
86271-24
86274-90
86284-80
86358-66
86390-20
86397-06
86400-34
86400-48
86403-12
86407-80
86408-10
86417-02
86423-10
86433-64
86442-24
86462-42
86471-00
86479-90

6.

86269-32
86259-77
86273-04
86268-33
86267-27
86273-16
86285-12
86357-73
8638848
86396-54
86400-59
86400-00
86403-31
86407-23
86408-94
86417-89
86424-60
86435-56
86438-77
86461-05
86470-50
86483-01

s.

+5-72
—4-23
+5-18
—0-15
—3-97
—1-74
+ 0-32
—0-93
-1-72
-0-52
+ 0-35
—0-48 .
+0-19
-0-57
+ 0-84
H-0-87
+ 1-50
+ 1-92
—3-47
— 1 37
—0-50
+3-11

Jamaica ....
New York . . .
Paris ...

Shanklin ....
Greenwich . . .
London ....
Arbury ....
Clifton ....
Altona . . . .
Leith . ...

Portsoy ....
Unst

Drontheim . . .
Harnmerfest . .
Greenland . . .
Spitzbergen . . .

From the column of " differences " we may obtain
the " probable error " of the vibrations at a single sta-
tion, and the proportion in which the probable error is
diminished as the number of stations is increased. Of
a single station the probable error is ls*7 ; of 10 stations,
Os<55.; and of 22 stations (the number in the table),

470 EDITOR'S NOTES.

08>36. The equatorial rate 86263S-60 consequently has a
probable error of + Os*36.

If attention be now directed to the " differences " at
the ten stations between the parallels of 40° and 60°
(which may be regarded as the representatives of the
middle latitudes in this investigation), we do not find,
either in the individual amounts, or in the systematic
occurrence of either + or — signs, the indications which
we should expect to find, if there were a notable discre-
pancy peculiar to those latitudes in the general form of
the ellipticity of the hemisphere. On the contrary,
the signs of opposite character are interspersed in a
manner which indicates that they are of an accidental
rather than of a systematic nature ; and in regard to
amount, the sum of the actual discrepancies at the ten
stations (giving their proper value to the signs) is 08*57;
whilst the (e probable discrepancy" of 10 stations (com-
puted from the discrepancies at the 22 stations,) is, as
above stated, Os<55.

If we combine the equatorial vibration 86263*60
successively and separately with the vibrations observed
at each of the 10 middle stations, and take the arith-
metical mean of the 10 results, we obtain from these
stations, so combined with the equatorial rate, '0051828
as the increase of gravity from the equator to the pole,
differing only -0000057 from the results of the 22

stations, and corresponding to the ellipticity L_ j

which, viewed as a partial result derived from the
middle latitudes, may be regarded as sensibly the same
as g 8*3. 4> already stated as the general result from the
equatorial and polar stations.

Having thus reasserted, and more fully justified, the

ON THE ELLIPTICITY OF THE EARTH. 471

symmetrical consistency of the ellipticity of the two
portions of the northern quadrant shown by the ac-
celeration resulting from the employment of Kater's
invariable pendulum, I may be permitted to rectify
an accidental mistake which occurs in p. 27 of the
original (p. 27 of this translation also), regarding the
the origination of the experiments in which I was myself
employed. The experiments of Captain Kater at the
principal stations of the British Trigonometrical Survey
were those which were undertaken, as stated by M. de
Humboldt, at the desire of H.M. Government, pursuant
to an address of the House of Commons, March 15th,
1816, which was moved by Mr. Davies Gilbert. This
undertaking was completed on June 19, 1819, when
Captain Kater presented the final account of his experi-
ments to the Royal Society. The subsequent extension
of the inquiry to stations including "the utmost accessible
distance on the meridian of a hemisphere," originated in
a proposition made by myself to the Board of Longitude,
and recommended by that board (which included
amongst its members several of the most eminent Fel-
lows of the Royal Society) to the favourable considera-
tion of H.M. Government. The object of the proposition
is stated in the preface of the work in which its accom-
plishment is recorded, viz., "to extend the suite of
stations to the equator on the one side, and to the highest
accessible latitudes of the northern hemisphere on the
other ; to multiply the stations at both extremities of
the meridian, so that, by their general combination, the
irregular influences of local density (which had prevented
any independent conclusion whatsoever, relatively to the
figure of the earth, being drawn from the experiments

472 EDITOR'S NOTES.

either of the French philosophers or of Captain Kater)
might mutually destroy each other, and the variations
of gravity due to the ellipticity alone be elicited ; and
to ensure the uniformity of procedure and strict com-
parability of the results at all the stations, by the unity
of the observer and the identity of the instruments. In
effect, to terminate the inquiry with the pendulum, —
either by obtaining decisively the result which it might
be capable of furnishing, — or by manifesting that no
decisive result whatsoever was attainable by it, even
under the most favourable circumstances of operation."
(Pend. Expts., 1825, Preface.)

I gladly avail myself of this occasion to recall, with
grateful recollection, the arrangements which existed, at
that period, for facilitating the communications between
Grovernment and those who by public estimation were
its most fitting counsellors, in matters appertaining to
the advancement of science and to the national duty
in its promotion. By means of the Board of Longitude,
a proposition from an individual, known only by the
desire he had shown to be employed on scientific under-
takings less suited perhaps to men of higher qualifica-
tions than himself, found a ready access to Grovern-
ment, when sanctioned by the approval of a body in
whose judgment the Government was accustomed to
place confidence.

When treating of the results of pendulum experiments
in the southern hemisphere (p. 28 of the Grerman, and
28 of the English edition), M. de Humboldt has omitted
to notice the experiments of Captain Henry Foster in

ON THE ELLIPTICITT OF THE EAETH. 473

the years 1828 — 1831, which, on account of the unity of
the observer, the identity of the instruments, the num-
ber of stations, and the extent of the arc which they
embrace, must undoubtedly be considered to hold the
first place in the determinations of that hemisphere. In
January, 1828, a communication was received by the
Royal Society from H.R. Highness the Duke of Clarence
(afterwards King William the Fourth), then Lord High
Admiral, intimating the favourable disposition of Go-
vernment to extend to the southern hemisphere the
investigation into the figure of the earth which had
been recently completed in the northern hemisphere ;
and requesting from the Royal Society such suggestions
as might be made the basis of the instructions to be
given to the officer who might be selected to execute
them. The reply was drawn up by a committee of the
Royal Society, of which Sir John Herschel was chairman,
and of which I was myself a member, and is preserved
in the archives of the Royal Society. Captain Foster,
who was appointed to conduct the experiments, had
been a midshipman with Captain Basil Hall, in the
" Conway," and afterwards with Captain Clavering, in
the " Griper," in which ship I was then visiting Spitz-
bergen and Greenland, and had recently received the
Copley medal of the Royal Society for his researches in
several branches of physical science made during the
arctic voyages of Sir Edward Parry in 1824 — 25 and
1827. The "Chanticleer" sloop of war having been
commissioned expressly for this service, Captain Foster
sailed from England in the spring of 1828, and in the
course of the two following years completed the experi-

474 EDITOR'S NOTES.

ments at the following stations, which are here ranged
according to their latitudes: —

Para

•

- 1°

27'

0" S.

Maranham

.

- 2

31

35 S.

Fernando de Noronha -

- 3

49

59 S.

Ascension

,

- 7

55

23 S.

Porto Bello

• m

- 9

32

30 N.

Trinidad

• m

- 10

38

55 N.

St. Helena

.

- 15

56

07 S.

Cape of Good

Hope

- 33

54

37 S.

Monte Video

•» m

- 34

54

26 S.

Staten Land

-

- 54

46

23 S.

Cape Horn

•

- 55

51

20 S.

South Shetland -

- 62

56

11 S.

To these twelve stations, of which ten are in south and
two in north latitude, we must add London and Green-
wich, where the pendulums were vibrated by Captain
Foster before his departure from England, and the
vibrations' repeated (in London) by Mr. Baily after the
return of the instruments at the conclusion of the voyage.
Having completed the experiments at Porto Bello, the
last of his pendulum stations, and whilst engaged in
measuring the difference of longitude across the isthmus
between Panama and Chagres, Captain Foster was acci-
dentally drowned in the river Chagres, on the 5th of
February, 1831. The papers containing his observations,
arranged but not calculated, were placed by the Admi-
ralty in the hands of Mr. Francis Baily, President of
the Koyal Astronomical Society, under whose directions
the necessary calculations were made, and the results
reported in a memoir forming the seventh volume of the
Transactions of the Royal Astronomical Society.

Captain Foster was furnished with two invariable

ON THE ELLIPTICITY OF THE EARTH. 475

pendulums of precisely the same form and construction
as those which had been employed by Captain Kater
and myself. Both pendulums were vibrated at all the
stations, but from some cause which Mr. Baily was un-
able to explain, the observations with one of them were
so discordant at South Shetland as to require their re-
jection. The acceleration computed from the mean of
the two pendulums at all the other stations, and from
the one pendulum at South Shetland, gives, according
to Mr. Baily's report, pp. 74 and 75, for the increase of
gravity from the equator to the pole, '0051924, and for
the ellipticity, g8*9.2. The very close agreement of this
result with that which had been previously stated from
the experiments in the northern hemisphere affords a
very strong probability that, in the conclusion common
to them both, we have a true measure of the ellipticity
of the earth derived by means of the pendulum, as well
as a highly interesting and important indication of sym-
metry in the two hemispheres.

There are two circumstances in particular which dis-
tinguish the double series of experiments discussed in
this note, and may claim for their result a higher degree
of confidence than has been accorded to the results of
experiments made under less favourable conditions.
The first is one which will be readily appreciated as soon
as stated, viz., the extent of the arc of the meridian on
which the stations of experiment are distributed — from
62° 46' S. to the equator, and from the equator to
79° 50' N. The second may perhaps be less readily
appreciated at the first view; but it is a condition,
nevertheless, of primary importance in results whose
value is to depend on their strict comparability with

476

each other — a condition which is best secured by the
employment of the same instruments and the same
processes of experiment throughout. From this con-
viction I have limited the notice which has been here
taken, both of Captain Foster's experiments and of my
own, to the results obtained with the pendulums of the,
same form and construction as those which had been
employed by Captain Kater, and with which the same
methods of observation which had been first practised
by him were carefully preserved. I was myself pro-
vided with two other invariable pendulums, differing in
some respects in form from Captain Rater's, and attached
to a train by which their vibration was maintained and
the number of vibrations in a day recorded ; and these
were employed as a wholly distinct mode of experiment
at the greater part of the stations of my equatorial and
polar series. The record of the results with these pen-
dulums is given in full in the volume of my Pend.
Expts., pp. 237 — 287, where their close accordance is
shown with those of Kater's. In like manner, Captain
Foster was provided, in addition to the two Kater's pen-
dulums, with a copper and an iron experimental pendu-
lum (both differing in form from Kater's), which were
used at several of the stations which he visited, and the
results are given by Mr. Baily in the report already re-
ferred to, where they are shown to accord sufficiently for
general confirmation, at the stations where they were
employed, with those of Kater. It is always a great
advantage to an experimentalist to be provided with
a second and wholly distinct process of experiment, —
though it may not be in all respects, perhaps, quite
so satisfactory as the one on which he principally

ON THE ELLIPTICITY OF THE EARTH. 477

relies; it inspires confidence if he finds the results
accordant, and it affords the opportunity if the results
should not be so, of endeavouring to clear up the
discrepancy on the spot. This was the chief purpose
which I contemplated in providing myself with the
attached pendulums ; and I found the full advantage of
the provision when, at stations nearly in the same lati-
tude with each other, and where, consequently, the rates
of the pendulums should theoretically have been nearly
the same, they differed to an amount of several seconds
in the day, the difference being shown by all the four
pendulums to nearly an identical amount ; leading to
the unavoidable conclusion that the discrepancy was not
due to errors of observation, but was the effect of natural
causes operating at the particular stations. In like
manner Mr. Baily, in calculating the results of Captain
Foster's experiments, appears to have derived from the
same circumstance the solution of the doubts, which he
is known to have entertained previously, of the true
causes of such discrepancies. In the report, p. 78, he
accompanies the table of comparative results by the re-
mark that the table " clearly shows that the vibrations
of a pendulum are powerfully affected, in many places,
by the local attraction of the substratum on which it is
swung, or by some other direct influence at present un-
known to us, and the effect of which far exceeds the
errors of observation. For it will be seen, from an in-
spection of the table, that all the pendulums tell (sub-
stantially) the same story at the same place."*

* It was at one time imagined that the angle which the plane in
which the pendulum vibrated made with the astronomical or mag-

478

In the tabular view which is given in page 469 of the
differences between the observed and computed rates,
Trinidad, Ascension, and Maranham are three stations
at which, in my experiments, the discrepancies were
most considerable in amount: —

At Trinidad the observed vibrations were in defect 3-97
At Ascension „ „ in excess 5'18

At Maranham „ „ in defect 4-23

Sum - - 13-38

For this reason it appeared to the committee of the
Royal Society, by whom suggestions were made for
Captain Foster's voyage, very desirable that the ex-
periments should be repeated at those stations. This
was accordingly done, and the results of the repetition
are stated in Mr. Baily's report in the table, p. 88. The
comparative results are as follows : — The pendulum,
which vibrated seconds in London, made at Trinidad
86267-27 vibrations according to Sabine, and 86267-24
according to Foster; at Ascension, 86273-04 Sabine,
and 86272-25 Foster; at Maranham, 86259-77 Sabine,
and 86258-74 Foster. At Trinidad and Maranham the
places of observation were identical ; at Ascension there
was a slight difference, — Captain Foster's pendulums
were vibrated in a house in the Barrack Square, mine in
a storehouse, within, I believe, a quarter of a mile's

netical meridians might influence its time of vibration. Experiment,
however, had already decided this question. At Sierra Leone, the
first station which I visited, the pendulums were successively
vibrated in planes at right angles with each other, with results so
nearly accordant as to be in effect identical. (Pend. Expts., pp.
15-16.)

ON THE ELLIPTICITY OF THE EARTH. 479

distance. The sum of the discrepancies of the two
observers from each other at the three stations is
ls-94; the sum of the discrepancies between the 06-
served and computed vibrations is, as above stated,
13s-38. Having been myself the person who specially
proposed this repetition, I may now permit myself
to refer to this statement, as an additional proof
that the discrepancies between the observed and com-
puted vibrations are due, in a far greater degree, to
local peculiarities, than to what may be more strictly
called errors of observation. The " probable error " at
a single station, calculated from the differences between
the observed and computed vibrations at the 22 stations
in p. 469, includes of course both station and observa-
tion errors; it is ls*7. The same computed from the
differences between the two observers at the three
stations above cited (regarding the arithmetical mean
between the observers as in each case the true result,
and half the difference between them as the error of
each), is about Os>3. The latter is the probable error of
observation ; the former the probable error of observa-
tion and station combined ; the error of observation is
therefore not a quarter as large as the station error.

The differences of opinion which once prevailed re-
garding the causes of such discrepancies have now passed
away ; and they are now universally admitted, I believe,
to be the effects of the unequal density of the superficial
strata of the earth. They were first recognised as such by
Captain Kater, in the Phil. Trans, for 1 8 1 9, pp. 424—426,
who drew from his own pendulum experiments at the
principal stations of the British Trigonometrical Survey
the conclusion that, " in passing through a country com-

480

posed of materials of various densities, the pendulum
may be expected to indicate such variations with very
considerable precision." The opinion thus expressed
was considered by me to be so strikingly confirmed by
the systematic character of the " anomalies" (as they
were then called) at the equatorial and polar stations of
my experiments in 1821-1822, as to justify the inference
(Pend. Expts. 1825, p. 341,) that "the scale afforded by
the pendulum for measuring the intensities of local
attraction appears to be sufficiently extensive tc render
it an instrument of possible utility in inquiries of a
purely geological nature. The rate of a pendulum
may be ascertained, by proper care, to a single tenth of
a vibration per diem ; whilst the variation of jate, oc-
casioned by the geological character of two stations, has
amounted, in extreme cases, to nearly ten vibrations per
diem : a scale of 100 determinable parts is thus afforded,
in which the local attraction dependent on the geolo-
gical accidents may be estimated." The inference thus
drawn was illustrated by a table (p. 338), exhibiting the
general nature of the surface-strata at the thirteen sta-
tions which I had visited in 1821-1822, with the excess
or defect at each station of the rate of the pendulum sup-
posed to be occasioned thereby, and the value correspond-
ing to the excess or defect in the scale of 100 parts; as
well as by a similar table (p. 345) for nineteen stations,
including those of Captain Kater. These were, I believe,
the first tables of the kind that were published. It is
chiefly to such local influences that we must ascribe the
discordances that have appeared in the ellipticity derived
from stations too few in number and too near each other
in latitude. If no irregular attraction occurred, the

ON THE ELLIPTICITY OF THE EARTH. 481

results computed from different portions of the meridian
should be the same. By multiplying the number of
stations the effects of the variations of superficial density,
sometimes in excess and sometimes in defect, tend to
counterbalance each other, and thus to diminish the
general amount of error ; whilst the increased extent of
the included arc gives to such error a diminished in-
fluence on the ultimate deduction.

The discordances which were formerly found in the
results of pendulum experiments, when the number of
stations were comparatively few, or when the arc which
they included was of small dimension, appear to have
produced in M. de Humboldt's mind an impression
(p. 29), that the pendulum is a less likely means of ob-
taining a well-assured conclusion respecting the figure
of our planet than the measurement of degrees. It is
not surprising, indeed, when we look back upon the
formidable array of distinguished geometricians and
astronomers who have taken part for more than a cen-
tury past in the measurement of degrees, that there
should be a bias in favour of the ultimate conclusions
from a method on which such a prodigious amount of
time, means, and skilled labour have been expended.
Still, there have not been wanting eminent persons who
have been of a different opinion ; and who, viewing the
discrepancies which have also presented themselves in
the ellipticity deduced from different measured arcs—
discrepancies arising from causes similar to those which
affect pendulum results — have anticipated that the pen-
dulum would eventually be found amongst the most
accurate means of determining the general configuration
of the globe. Happily, we have arrived at the time

VOL. IV. I I

482

when it is no longer necessary to discuss the claim of
either of the methods to have its result accepted in
preference to that of the other, since they are now found
to unite in a conclusion which, if not absolutely identical,
is so nearly so as to leave scarcely anything more to be
desired. The calculations of the Russian arc, extending
from the North Cape to the Black Sea, and the revision
' of the Indian arc with the comparison of its standard
scale with that of the Russian arc, which are referred to
in the text (p. 21) as having been in progress when that
part of the volume was written, have since been com-
pleted ; and although the details are not yet published,
it has for some time past been known, on the authority
of M. Struve, that the result of this combination is to
manifest a much greater compression than had pre-
viously been deduced by Airy and Bessel in 1830, 1837,
and 1841 from all the arcs then at their command. It
is now known that the ellipticity derived from the
Russian and Indian arcs is between ^Y and -^j ; and
from the magnitude of those arcs (the Russian exceed-
ing 25°, and the Indian exceeding 21°) and the great
meridional extent embraced between them, it is not
likely that this conclusion will be disturbed by exten-
sions which yet remain to be completed of arcs of less
magnitude, which, being also in intermediate latitudes,
are of less importance in regard to the ellipticity.

It may be worth while to cast a retrospective glance
on the progressive assimilation of the ellipticity deduced
from the measurement of degrees, as these measures have
increased in number and magnitude and have brought
i-nto- comparison more distant portions of the meridian,
to the ellipticity derived from the pendulum. Without

ON THE ELLIPTICITY OF THE EARTH. 483

laying undue stress on -j-^, the value assigned by M. La-
place in his " Exposition du Systeme du Monde, 1799,"
we find in the second volume of Biot's " Astronomic
Physique," published in 1810, y^Ts staked, as the
precise result of a careful combination of all the arts of
the meridian which had been measured up to that date.
In 1830, Mr. Airy, after a careful consideration of all
the arcs which had been then measured, estimated their
most probable result at 29g.33. (Encyc. Met. Figure of
the Earth.) In 1837, M. Bessel, whose acquirements and
habits specially qualified him for such a task, undertook
the revision and rectification of the earlier arcs, and
their combination with three additional arcs which had
been then recently completed. The general result was
,,—ry. The extension of the Eussian, and the later
revision of the Indian arc, have made another step in
the value of the denominator of the fraction, of nearly
the same amount as the preceding step between the
conclusions of M. Biot and of MM. Airy and Bessel. The
difference between the ellipticity which is now found
to result from the measurement of degrees, and that of
the pendulum, is reduced to 3 in the denominator
of the fraction which denotes the amount of the com-
pression ; or to about a fifth part only of the change
which additions and rectifications in the measures of
degrees have made since the early part of the century.
Viewing the obvious tendency of the two methods—
the pendulum and the measurement of degrees — to
unite in one and the same conclusion, we may surely
permit ourselves to regard the still subsisting very small
difference between them as comparatively, if not wholly,
insignificant. In regard, however, to this really slight

II 2

484 EDITOR'S NOTES.

remaining difference, I own that I am individually
inclined to give the preference to the pendulum result.
This preference may, no doubt, be in part owing to a
natural bias in favour of what I have been myself
engaged in, and, so far, be entitled to the less weight :
yet it is to be remembered that the pendulum result
has the advantage of nearly ten degrees of greater
meridional extension than the measured arcs, and that
the latter have been limited almost exclusively to the
one (the northern) hemisphere, whilst pendulum ex-
periments have been extended over both hemispheres
with almost concurrent results.

In conclusion, there is one advantage which cannot
be denied to the pendulum over the measurement of arcs,
namely, the far less expenditure of time, means, and
skilled labour which it requires. The time that inter-
vened between the proposition to give the pendulum
experiments the widest practicable extension between the
eo^ator and the pole, — and the completion of that un-
dertaking and the publication of the results, — was barely
five years. The experiments and the calculations were the
work of a single person ; and the conclusion from them,
widely as it then differed from received opinions, has
since stood as the mark towards which the subsequent
results of the measurement of degrees have progressively
approximated ; and which, after a lapse of more than
thirty years, bids fair to prove the true measure of the
general figure of our planet.

ON THE MAGNETIC DISTURBANCES. 485

II. — On the Magnetic Disturbances.

THE term-observations of the Grermaii Magnetic Asso-
ciation, superintended by MM. Gauss and Weber in
1835 and the following years, had shown that the
occasional and apparently irregular disturbances of the
declination magnet (to which element the German
term-observations were principally confined) took place
almost always contemporaneously, over the large portion
of the interior of the European continent comprehended
by their stations; and the numerous and rapidly vary-
ing fluctuations in the direction of the magnet which
were observed to take place during a disturbance at one
station, were found to occur simultaneously at other
stations with an accordance admitting of no mistake.
These facts, established by multiplied observations, gave
a great increase to the interest with which the minor
magnetic variations had been previously regarded, and
caused a general desire to be felt that the investigation
should be extended to more distant parts of the globe ;
and that, in particular, the questions of the contem-
poraneous occurrence of the disturbances, and the simul-
taneity of the fluctuations which they occasioned, should
be examined over a wider field than that of the German
Association, and should embrace different continents
and different hemispheres.

But whilst the public attention was chiefly fixed on
what were undoubtedly the most prominent and the
most striking conclusions established by the extensive

486 EDITOK'S NOTES.

system of co-operative investigation which had its head-
quarters at Grottingen, there were others also which,
although they attracted less notice at the time, were not
less apparent upon a close examination of the simul-
taneous observations at distant stations. By a careful
examination of the diagrams in which these were repre-
sented in the Resultate aus den Beob. des magnetischen
Vereins, it could not fail to be perceived that (after due
allowance was made for the small differences at the
several stations of the absolute horizontal force of the
earth, which is the antagonistic of the disturbances),
the disturbing causes, whilst acting simultaneously, pro-
duced more extensive deflections at certain stations than
at others, indicating thereby that the energy of the dis-
turbing force itself was subject to variations due to un-
known peculiarities. Examples were not infrequent also,
in comparing the movements of the declination magnet
at two stations, of a difference in the proportional mag-
nitude of successive deflections, ona deflection being
greatest at the one station, and another deflection
greatest at the other station. Though almost always
exhibiting the same feature of simultaneity, the deflec-
tions showed considerable variety, and apparent irregu-
larity, in their extent. The conclusion seemed almost
unavoidable, and appears accordingly to have been
entertained by M. Gauss himself, that more disturbing
forces than one were contemporaneously in action ; that
these were, possibly, quite independent of each other,
and might have very different sources; their effects
being intermixed in very dissimilar proportions at places
differently situated in relation to the position and dis-
tance of their respective sources. The disentanglement

ON THE MAGNETIC DISTURBANCES. 487

of the complications produced by this presumed super-
position of phenomena, — the separation of the indi-
vidual forces of which the fluctuations are the compound
result, — and the assignment of the source and measure
of each, — were viewed by that eminent geometrician as
"triumphs of science" which, though t( incontestably
very difficult of accomplishment," might not be unat-
tainable by perseverance in the course of inductive
investigation which had been so happily commenced.
(Eesultate, 1836, pp. 99—100.)

The scheme of the British colonial observatories,
adopted by the British Government on the joint recom-
mendation of the Royal Society and of the British Asso-
ciation for the advancement of science, was much more
comprehensive than that of the Grerman Association, inas-
much as it included the investigation of the laws of all the
magnetic variations, secular, periodical, and occasional,
sensible al* the surface of the globe ; and embraced as
distinct subjects of inquiry the phenomena of each of
the three magnetic elements, the declination, the incli-
nation, and the intensity of the magnetic force, — the
whole phenomena, in fact, of the magnetic direction and
force, as they might be severally affected by the different
variations. The occasional disturbances (the subject of
this note) still preserved, however, a prominency in the
order in which the investigations were to be made; from
the circumstance, that the progressive and periodical
variations appeared to be so mixed up with the in-
fluences of the transient and occasional, that a know-
ledge of. the laws and relations of the latter, whereby
their effects might be eliminated, was deemed to be
indispensable towards a correct investigation and analysis

488 EDITOR'S NOTES.

of the former. The study of the occasional disturbances
was therefore taken up at the point to which it had
been carried by the German Association. The ques-
tions,— of general contemporaneous occurrence, — and of
the degree of simultaneity and accordance in individual
fluctuations of the disturbed magnets, — were examined
at observatories established in Africa and America, as
well as in Europe, in the northern and southern hemis-
pheres, and in the tropics, by observations of the three
elements at short intervals, at preconcerted instants of
absolute time, as well as generally on all occasions when
fluctuations of more than usual amount presented them-
selves in the course of the regular hourly registration.
The result of this examination was conclusive in refer-
ence to the contemporaneous occurrence of the phe-
nomena in all parts of the globe. Not only were the
same days characterised in the records of the different
observatories as being "days of unusual disturbance,"
but even the degree assigned to the disturbance, as
"excessive," "great," or "moderate," was usually the
same at stations most distant from each other. The
instants of extreme disturbance in one of the elements
in one part of the globe were frequently as nearly iden-
tical as the nature of the observation permitted, with
those of the same or another element in a very distant
part ; and the culminating points of minor fluctuations
as frequently exhibited a degree of accordance cha-
racteristic of the existence of general affections, extend-
ing simultaneously over the whole surface of the globe,
but ' influencing the magnetic elements variously, ac-
cording, possibly, to the geographical position of the
station, or to its distance and direction from the locali-

ON THE MAGNETIC DISTURBANCES. 489

ties which may eventually be found to be those of the
sources of disturbance. The correspondence which had
been manifested in the fluctuations of the declination
in Europe was also found to prevail in much the same
degree when the term-day curves at different stations
in Canada and the United States were compared inter
se; but in a far less degree when the simultaneous
curves of the two continents were placed in comparison.
Indications indeed were not wanting of the participation
of both continents in disturbances having a common
origin; but additional and still stronger evidence, of
dissimilarity in the particular effects produced by the
disturbing action in different localities, gave weight to
the earlier inference of an admixture of forces, acting
simultaneously and occasioning complication, for the
disentanglement of which some new process of investi-
gation appeared to be required. Now it was obvious
that if more forces than one were in general action at
the same time, a properly directed analysis of the dis-
turbances at a single station might separate their effects,
and exhibit their respective laws more simply and
readily, than when additional complexity was introduced
by having to consider at the same time the different
action of the same forces at different places. In such
an inquiry the first questions which would present them-
selves for solution would be, whether any periodical
influence were discoverable in the mean effects of phe-
nomena which were themselves of occasional, and in a
certain sense, irregular occurrence ; and if so, whether
the different magnetic elements had the same or dif-
ferent periodic laws ; and whether, in each element taken
separately, the fluctuations which were observed to take

490 EDITOR'S NOTES.

place on opposite sides of the mean or normal values
had the same or different periodic laws. For this inves-
tigation, the hourly observations, which had been main-
tained with steady perseverance for several years at the
British Colonial Observatories, appeared likely to furnish
a suitable provision, if a selection were made from them
of a sufficient body of the particular class of phenomena
to enable their laws to be studied. As the magnitude
of deflection was the only known criterion of a dis
turbance of the class which had been termed "irre-
gular," a number of the largest deflections (reckoned
from the mean or normal values in the same month
and at the same hour) were separated from the whole
body of the observations for the proposed examination.
Care was taken that the amount of deflection, which
should cause an observation to be regarded as "dis-
turbed," should be on the one hand so low as to sepa-
rate a sufficient number of disturbed observations for the
deduction of mean effects, whilst, on the other hand, it
should be so high as not to include irregularities of the
ordinary regular and well-known variations : and that
the same disturbance-value should be taken throughout,
in order to insure a just comparison between different
months and different years. The observations so sepa-
rated were formed into tables both of the numbers and
of the aggregate values of the deflections (reckoned from
the normals) according to the years, months, and hours,
of their occurrence ; and the mean annual, monthly,
and daily numbers, and aggregate values, being severally
taken as units, the relative proportions of frequency
and of amount of disturbance, in the several years,
months, and hours, were obtained. The aggregate

ON THE MAGNETIC DISTURBANCES. 491

amounts of deflection were also divided into " deflections
on either side of the normal," and treated in the same
way. The variation in relative frequency and amount
in the respective periods being the object of this inves-
tigation, the absolute numbers and values of the sepa-
rated observations would have no other importance
than as evidencing that they were sufficiently large to
justify the deduction of the relative proportions. The
numbers and values would, of course, be larger or smaller
according as the amount of deflection taken as the mea-
sure of a disturbed observation were lower or. higher ;
but it was soon found on trial, that there were always
certain moderate limits within which this amount might
be taken indifferently without producing any material
alteration in the relative proportions. At stations where
the energy of the disturbing force was great in com-
parison with the deflections produced by the ordinary
and regular variations, a larger body of disturbed ob-
servations could be safely separated: but in no case
could the separated values be regarded as furnishing a
measure of the whole effects of the occasional disturb-
ances ; inasmuch as, until some other criterion than that
of magnitude is known, by which an effect of the par-
ticular causes now under notice can be discriminated,
there must always be supposed to be minor effects of the
same causes remaining amongst the unseparated body.

By pursuing this method of investigation with the
observations that had been received from the observato-
ries of Toronto, St. Helena, and Hobarton (stations widely
distant from each other on the globe), the same conclu-
sions were obtained from each, viz. that the disturb-
ances are governed, in the frequency of their occurrence,

492

and in the magnitude of their mean effects, by several
periodical laws of a highly systematic character, admit-
ting of well-assured determination. Perhaps the most
important of these periodical laws is the one which has
been honoured by M. de Humboldt's notice more than
once in the present volume, because its discovery first
established the connection of the terrestrial magnetic
variations with the periodical variations in frequency
and amount of the solar spots, thus affording a more
certain evidence than any which had preceded it, of a
direct magnetic influence emanating from the central
body of our system, and producing phenomena at the
surface of our planet, in the production of which it
was scarcely possible to imagine that the solar magnetic
influence acted through the instrumentality of varie^
tions of temperature. The decennial period, as it has
been termed (though some uncertainty still prevails
as to its precise length), is apparently a solar period,
inasmuch as we perceive it to form a recurring cycle in
the physical appearance of the solar disk, whilst on our
earth it has no other physical manifestation or accom-
paniment than those magnetic variations, which in other
ways also (as in their annual and diurnal periods) mani-
fest their dependence upon the sun. The epoch of
greatest obscuration in the solar disk by the dark spots,
is characterised on our globe by a maximum prevalence
of the magnetic disturbances, by their greater frequency,
duration, and occasional magnitude, and consequently
by the greater amount of their aggregate effects ; whilst
the epoch of minimum in the solar obscuration is analo-
gously characterised by a corresponding minimum in
the magnetic affections. The decennial period is shown

ON THE MAGNETIC DISTURBANCES. 493

separately by each of the three magnetic elements, and
has been evidenced alike wherever observations have
been maintained for a sufficient time to include the two
opposite epochs of the period, and have been submitted
to a suitable process of examination.

The primary connection of the disturbances with the
sun is evidenced also by the discovery, (which has been
another result of the mode of investigation referred to),
that their mean effects, in all parts of the globe and in
each of the elements, are found to be invariably governed
by periodical laws, whose period is a mean solar day.
This is the one feature which is constant, amidst diver-
sities in the hours of maximum and minimum of the
different elements at the same station, and of the same
element at different stations, — diversities which fully
account for the absence of absolute correspondence in the
simultaneous movements of the magnetometers at places
distant from each other, and harmonise with the con-
jecture of the contemporaneous existence of more than
a single force. It is in this view especially to be noticed
that, at the same station, the disturbances which pro-
duce easterly deflections of the declination, and those
which produce westerly deflections of the same, have
distinctly different laws ; as have also the disturbances
which increase and those which decrease the inclination,
as well as those which increase and those which decrease
the intensity of the magnetic force. The laws are in
all these cases highly systematic, the hours of maximum
and minimum well marked, and the progression regular.
There are thus, at each station, six distinct phenomenal
laws of this particular class of phenomena coexisting,
and necessarily producing much complexity by their

494 EDITOR'S NOTES.

combination ; and when to this complexity at one
station is superadded the additional complication of
the diversity of the action of the same forces at different
stations, an amount of intricacy is produced, which has
caused the earlier method of investigation, by the com-
parison of term-day observations, to yield latterly so
little fruit in proportion to the labour which has been
bestowed in the accumulation of materials. By the
substituted method, and by taking the stations sepa-
rately, the intricacies are unravelled ; and each of the
six laws is found to admit of separate determination.
Although the three stations already named, at which
this has been accomplished, are too few and too far
between to furnish an adequate basis for generalisation,
yet, the measure of success already obtained gives us
full reason to believe, that by perseverance in the path
which has been opened, we have a fair prospect of dis-
cerning the mutual relations which cannot fail to belong
to forces having the same origin, as well as of distin-
guishing more clearly than we are at present able to do
between forces which have distinct sources ; and thus of
completing the " triumph of science " which was antici-
pated by the great geometrician.

Already the very recent addition of the disturbance
laws of but a single element (the declination) at a new
station, Point Barrow (Phil. Trans., 1857, art. xxiv.), has
made known a mutual relation, of a most remarkable
and suggestive character, existing between the pheno-
mena at Toronto and Point Barrow, — the laws of the
easterly deflection at the one station being found to
correspond at the same local solar hours with those of
the westerly deflections at the other station, and vice

ON THE MAGNETIC DISTURBANCES. 495

versa. (There is about six hours" difference in the ab-
solute time corresponding to the same solar hours at the
two stations.) The great disproportion also in the mag-
nitude of the disturbances at Toronto and Point Barrow
is most conclusive evidence of the greater energy of
the disturbing forces at some parts of the globe than at
others ; and leads to the natural inference that at Point
Barrow we are in close proximity to one of the so.urces
from which their actions emanate.

The extraordinary manifestation of concomitant au-
roral phenomena, and the insight which we have gained
thereby into the horary laws of their occurrence, and
into the relation of these laws to those of one in parti-
cular of the disturbance variations, unite in indicating
the neighbourhood of Point Barrow as one of the most
important localities on the globe for the further prose-
cution of these researches.

496 EDITOR'S NOTES.

III. — On the solar-diurnal variation of the Magnetic
Declination.

UNDER the ordinary name of "diurnal variation" are
usually included several independent variations, distin-
guishable apart by their respective hours of maxima and
minima, and by the characters of their progressions,
but which unite in one common feature and are thereby
traceable to one common primary cause ; the common
feature being that, in all, the period of variation is a
mean solar day. Our first knowledge of these variations
is obtained by observing them in their conjoint or
accumulated effects, viz. in the changes of direction
which they produce in a freely suspended magnet at the
different hours of the day and night. It is the business
of the analyst, by various modes of grouping and com-
bining 'the observations, to distinguish from each other
the effects which proceed from different sources, and as
the first step towards a knowledge of the physical caus-
ation by which such dissimilar effects are made to pro-
ceed from the same primary source, to ascertain their
several phenomenal laws. The investigation which has
this purpose in view has already made some progress ;
and its advance may perhaps be aided by an endeavour
to classify the natural facts of which a knowledge has
thus been acquired. The endeavour must necessarily
be imperfect, because our knowledge is still but partial
and incomplete, as well as of very recent acquisition ;
and because we have as yet no true theoretical perception

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 497

of the agency by which any of the natural effects are
produced. It may tend also to a better appreciation of
the difficulties to be overcome, in obtaining a scientific
knowledge of the phenomena which present themselves
to the magnetic observer at the first outset, (viz. the
changes which he perceives to take place in the compass-
needle in the course of a single day), that it should be
shown what a variety of effects are combined in those
changes sometimes thought to be so simple. We may
commence the classification by describing the principal
characteristic features of the mean diurnal variation ; or
the result we obtain when we take an average of the
diurnal variation observed on every day of the year,
and thereby eliminate for the moment any systematic
differences which may belong to different portions of
the year.

§ 1. Mean diurnal variation.

When proper precautions are taken to protect a decli-
nometer from casual and non-magnetic disturbances,
and when means are taken for every hour of all the
observations made at that hour throughout the year, it
is found that there are certain stations on the globe where
these means show scarcely any notable difference in the
direction of the magnet at any hour of the day or night.
In conformity with the terms frequently employed of
" north and south magnetic hemispheres," such stations
may be magnetically termed " equatorial stations." In
receding from them, either to the north or to the south,
indications begin to appear of a systematic horary dif-
ference, i. e. a diurnal variation, in the declination, and

VOL. IV. K K

498

these progressively augment in arc-value, until in the
middle latitudes of both magnetic hemispheres the
extent of the variation, or the difference between the
extreme easterly and westerly values in the course of
twenty-four hours amounts, on the annual mean, to
nine or ten minutes. The deflections, or departures
from the mean declination are greatest during the hours
of the day; a circumstance which does not require,
however, the actual presence of the sun above the
horizon, as it takes place also in the arctic circle during
the polar winter. The extreme deflections occur, in the
middle latitudes of both hemispheres and in all meri-
dians, at nearly similar hours of solar time ; but with
the characteristic distinction, that, (speaking always
of the north end of the magnet,) in the northern
hemisphere the extreme easterly deflection takes place
in the forenoon, and the extreme westerly in the
afternoon ; and conversely, in the southern hemi-
sphere, the extreme westerly deflection in the fore-
noon and the extreme easterly in the afternoon. The
hours of the extremes are, for the forenoon, from 7
to 10 A.M., and for the afternoon from 1 to 3 P.M.: the
change of direction is most rapid from 10 to 11 A.M.
The hours of extreme deflection, or, as they are fre-
quently called, the " turning hours," do not appear to
depend, as some physicists have supposed, on the angle
which the magnetic direction may make with the astro-
nomical meridian (viz. on the declination); as the
hours are approximately the same in places which
differ very widely from each other in that respect. For
example, at Dublin, where the declination is 29° west;
at Prague, where it is 17° west; at Toronto, where it is

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 499

2° west; and at Point Barrow, where it is 41° east;—
at all these places the westerly extreme takes place
within a few minutes of 1 P. M., and the easterly extreme
between 7 and 8 A.M., although the extreme difference in
the declination is 70°, which is equivalent to a difference
of between four and five hours in the magnetic azimuth
of the sun at 1 p. M. From the great distances apart of
these four stations it may not be unreasonable to infer,
that the turning hours in which they all agree may be
found to be approximately the same, (apart from the
effects of the occasional disturbances), throughout the
middle latitudes of the northern hemisphere. When we
examine the corresponding phenomena of the southern
hemisphere, however, we find a remarkable difference
in the turning hours of the middle latitudes, which
(though not inconsistent with what is stated in the
earlier part of this note, viz. that the hours in both
hemispheres are nearly similar) seems to indicate a
systematic hemispherical difference of fully an hour and
a half in the times of occurrence both of the forenoon
and afternoon extremes. Hobarton is the station, in the
southern hemisphere, where the hours of extreme de-
flection have been most securely ascertained, on the
extensive basis of eight years of hourly observations.
These have given between 9 and 10 A.M. for the time
of the forenoon extreme, and between 2 and 3 P.M. for
that of the afternoon extreme. The turning hours ob-
served by Sir James Ross in his Antarctic Expedition, at
the Auckland Islands, at New Zealand, and at Sydney
are in accordance with those at Hobarton. Although
the hourly observations at Sir James Ross's three stations
were continued but for a few weeks at each, the results

500 EDITOR'S NOTES.

which they have thus yielded are most valuable when
taken in conjunction with the conclusions from the
Hobarton observations; which if they had stood alone
might have been deemed indicative of local peculiarity,
rather than as affording a distinct hemispherical type.
In the present state of our knowledge, or rather in our
present deficiency of knowledge, no physical explanation
of such a systematic difference of the turning hours in the
two hemispheres readily presents itself; and we must
therefore consider it as particularly desirable that
further investigation should be made of the correspond-
ing phenomena in other parts of the middle latitudes
of the southern magnetic hemisphere.

Besides the deflections which occur during the hours
of the day there are others, though of much minor
extent, which occur during the hours of the night.
There is reason to believe that these are in great mea-
sure if not altogether produced by the occasional dis-
turbances; which, as stated in the preceding note, are
now known to be governed in their mean effects by
periodical laws, amongst which is one which has for its
period a solar day, and has different hours of maximum
and minimum from those which have been described as
belonging to the regular diurnal variation. But these
effects, as well as peculiarities which have been occa-
sionally found in the diurnal variation observed in the
higher latitudes of the northern hemisphere, and which
there is reason to believe are also traceable to the effects
of the occasional disturbances in the regions where those
effects are the most largely developed, will be more
advantageously treated of in a subsequent section.

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 501

§ 2. The semi-annual inequality.

The diurnal variation described in the preceding sec-
tion is that which we find when we take an average of the
whole year, and corresponds with what we may suppose
would actually take place throughout the year if the
sun's path were always in the plane of the equator.
But observation has made known to us the existence of
a semi-annual inequality in the diurnal variation,
having opposite phases according as the sun has north
or south declination; having consequently its turning
epochs about the times of the solstices, and disap-
pearing, as one phase passes into the other, about the
times of the equinoxes. As far as can be judged
from the stations where the facts have been most satis-
factorily examined, which are chiefly equatorial and
middle latitude stations, the semi-annual inequality is a
general phenomenon, and nearly uniform in character
and amount at all stations. Its principal effect takes
place during the hours of the day, the extreme deflec-
tions occurring about 7 or 8 A.M., and from 1 to 2 P.M. ;
the north end of the magnet being deflected in the
forenoon to the east when the sun is north of the
equinoctial, and to the west when he is south of
it; and, conversely, in the afternoon to the west
when the sun is north, and to the east when he is
south of the equinoctial. In other words its effect
is, — when the sun has north decimation to cause the
north end of the magnet in both hemispheres to be
more to the east in the forenoon, and more to the west
in the afternoon, than it would otherwise be, — and

502 EDITOR'S NOTES.

when the sun has south declination to cause the north
end of the magnet in both hemispheres to be more to
the west in the forenoon, and more to the east in the
afternoon, than it would otherwise be. The range of
deflection from one solstice to the other is about 3' or
4' at the extreme or turning hours ; and the daily range
from the forenoon to the afternoon is throughout the
year also about 3' or 4', except about the equinoxes when
the opposite phases pass into each other. The transition
from the one phase to the other takes place rapidly,
and is nearly completed a few days after each of the
equinoxes.

We may now consider the general effect of the super-
position of the semi-annual inequality upon the mean
diurnal variation stated in the preceding section. In
the middle latitudes of the northern (magnetic) hemi
sphere, the eastern extreme of the mean diurnal varia-
tion has been described as being reached about 7 or 8
A.M., and the western extreme about 1 or 2 P.M. ; the
mean range during the year being about 9' or 10':
whilst in the middle latitudes of the southern (mag-
netic) hemisphere, the contrary extremes are attained
nearly at the same hours ; and the mean range is the
same, or nearly so. Now, as the range of the semi-
annual inequality is less than 9' or 10', its superpo-
sition upon the mean diurnal variation in the middle
latitudes in either hemisphere does not alter the cha-
racter of the forenoon or afternoon extremes in any
part of the year, but it does alter their amount.
During the greater part of the half year in which the
sun has north declination, the semi-annual inequality

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 503

described in this section has the effect in the northern
hemisphere of increasing both the forenoon easterly and
the afternoon westerly deflections, and in the southern
hemisphere of decreasing both the forenoon westerly
and the afternoon easterly deflections — and during the
greater part of the other half year in which the sun has
south declination, the semi-annual inequality has the
opposite effects, i. e. in the northern hemisphere, it
decreases both the morning easterly and the afternoon
westerly deflections, and in the southern hemisphere, it
increases both the morning westerly and the afternoon
easterly deflections. The apparent complexity of the
phenomena which we have described is occasioned by
the remarkable difference which takes place in the mode
in which the sun's influence is exercised in producing,
1. the mean diurnal variation, and 2. the semi-annual
inequality of the same. In the first the directions of
the deflection are uniform throughout the year in the
middle latitudes of the one hemisphere, and (although
opposite) are also uniform throughout the year in the
middle latitudes of the other hemisphere ; whilst in the
semi-annual inequality the directions of the deflection
are uniform in the two hemispheres, but opposite in the
two half years. Thus in each hemisphere the semi-
annual deflections concur with those of the mean di-
urnal variation for half the year, and consequently
augment them ; and oppose, and consequently diminish
them in the other half year.

We have next to consider the effect of the super-
position of the semi-annual inequality at the magnetic
equator and adjacent stations, using the term "mag-

504 EDITOR'S NOTES.

netic equator" in this case to denote a line passing round
the globe, and dividing it into two hemispheres which
are distinguished apart by the mean diurnal variation
having opposite deflections at the same hours throughout
the year. On the dividing line there is no mean diurnal
variation ; but it is by no means a line without diurnal
variation, for in each half year the alternate phases of
the semi-annual inequality constitute a diurnal varia-
tion of which the range in each day is about 4 minutes,
and which takes place every day in the year except
about the equinoxes ; the march of the diurnal varia-
tion being from east in the forenoon to west in the
afternoon when the sun has north declination, and from
west in the forenoon to east in the afternoon when he
has south declination. As the magnetic equator is re-
ceded from on either side towards the middle latitudes,
the mean diurnal variation characteristic of the hemi-
sphere becomes more and more sensible, producing, with
the deflections of the semi-annual inequality, compound
phenomena, in which the character of the semi-annual
inequality, (consisting of opposite deflections at the
same hours in opposite parts of the year,) is preserved,
until the amount of deflection at the turning hours of the
mean diurnal variation preponderates over that of the
semi-annual inequality : when the phenomena pass from
the character of equatorial to that of middle latitude
stations. St. Helena is an example of the equatorial
phenomena; Toronto of the phenomena described as
being those of the middle latitudes of the northern
hemisphere; and Hobarton of those of the middle lati-
tudes of the southern hemisphere.

The description which has been given of the pheno-

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 505

mena of the semi-annual inequality is in no way in-
consistent with the ingenious hypothesis suggested by
the late Dr. Langberg for its explanation (Proc. Eoyal
Society, vol. vii. p. 434). The epoch of the transition from
the one semi-annual phase to the other certainly takes
place about the time of the equinoxes, but it cannot be
said that observation has yet denned the precise day of
the transition at either equinox.

The important question next presents itself, by
what other geographical or magnetical peculiarities is
the line distinguished, which we may regard as the
zero-line of the mean diurnal variation, and where
consequently the semi-annual inequality constitutes the
sole subsisting diurnal variation? The places which
first made known to us these characteristic phenomena
of a magnetically-equatorial station were St. Helena
and the Cape of Grood Hope. Both are situated near
the line of least magnetic force on the surface of the
globe. This is a line encircling the globe, passing
through and connecting the points in each terrestrial
meridian where the magnetic force is least in that meri-
dian : it is not therefore itself an isodynamic line, but
divides the surface of the globe into two hemispheres, in
each of which the magnetic force progressively increases
on either side of the dividing line, until the isodynamics
form lemniscates around the two foci of maximum
force in each hemisphere. In this respect it has a
claim to be considered as a magnetic equator ; and its
claim will be greatly strengthened if it be identified, as
seems probable, with the zero -line of the sun's mean
diurnal influence on the globe. It would be super-
fluous to dwell on the obvious theoretical importance of

506 EDITOR'S NOTES.

establishing decisively, by observations at a sufficient
number of well-selected localities, whether this line of
least force, or some other line which according to va-
rious hypotheses might be deemed to have a preference,
be really the zero line of the mean diurnal variation; viz.
the line in which, apart from the semi-annual effect pro-
duced according as the sun has north or south declination,
the diurnal variation disappears, and in which the semi-
annual effect also disappears about the times of the
equinoxes, — but only then. Those who incline to ascribe
all the minor magnetic variations, and prominently
amongst them the diurnal variation, to thermic agency,
or to thermo-electric currents generated on the surface
of the globe or in the higher regions of the atmosphere,
might, according to distinctions in their particular views,
be disposed to prefer either the terrestrial equator, or
some variable line, such as "the parallel of latitude
which has the sun in its zenith," as the most probable
line from which effects should commence, which are
supposed to proceed from the more heated to the
less heated portions of the globe. Amongst those
who, on the other hand, might deem it more pro-
bable that ^effects which are in themselves magnetic,
might have a reference in their distinguishing pheno-
mena to characteristic lines of the earth's magnet-
ism, there would doubtless be many who would be
disposed to give a preference to the line of no dip,
so frequently termed the magnetic equator. In such
important questions we naturally desire indisputable
evidence; but this we scarcely as yet possess, though
the facts already known appear to point most distinctly
to the line of least force as meriting a decided preference.

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 507

The station, St. Helena, which first placed in a clear and
unmistakable point of view the peculiar features of the
semi-annual inequality, and the at least approximate
absence of any other diurnal variation, is in 15° 56' south
latitude, and has a magnetic dip of 22°. The distance,
therefore, is considerable, both from the terrestrial
equator and from the line of no dip ; and in respect to
the sun's vertically, he is in the zenith of St. Helena in
the first week of November and the first week of
February, whereas the equinoxes are the epochs when
the opposite phases of the semi-annual inequality pass
into each other, and all diurnal variation disappears.
On the other hand, St. Helena is situated in the close
vicinity .of the present position of the line of least force,
which, in consequence of the great distance apart (in
that quarter of the globe) of the two foci of maximum
force in the southern hemisphere, has a large southerly
inflection, and passes between St. Helena and the
Cape of Grood Hope, but somewhat nearer to St. Helena.
The Cape of (rood Hope furnishes still stronger evi-
dence to the same effect: the phenomena are similar,
indeed nearly identical with those at St. Helena, mark-
ing the Cape distinctly as a magnetically-equatorial
station, whilst its geographical latitude is 33° 56 ', and
its dip 53° (very distant therefore from the line of no
dip), and the sun is not vertical in any part of the year.
A very instructive contrast is presented when we com-
pare the diurnal variation at Algiers and at the Cape of
Good Hope ; the two stations are at opposite points of
the same continent, Algiers in 36° north, and the Cape in
34° south latitude, and are therefore nearly equidistant
from the terrestrial equator : and as the dip is 57° north

508

at Algiers, and 53° south at the Cape, they are also both
very distant (and nearly equidistant) from the line of no
dip. But while their situation is thus nearly symmetrical
in regard to distance from the equator and from the line
of no dip, it is much otherwise in respect to the line of
least force, which enters the African continent only a little
to the north of the Cape of (rood Hope, and at a very
much greater distance from Algiers. Now, at the Cape,
the semi-annual inversion of the hours of extreme
easterly and extreme westerly elongation is a very
decided feature, scarcely less so than at St. Helena.
When the sun has south declination, the extreme
westerly elongation is in the forenoon, and the extreme
easterly in the afternoon ; and when the sun has north
declination, the converse takes place both in the fore-
noon and afternoon. The Cape thus possesses this
peculiar character of a magnetically equatorial station.
At Algiers, on the other hand, the extreme easterly
elongation is, throughout the year, in the forenoon, and
the extreme westerly in the afternoon. (Aime : Explor.
Sci. de 1'Algerie, 1846, t. ii. p. 218.) Algiers has not,
therefore, this character of a magnetically equatorial
station ; whilst the Cape has it. Again, at the Cape,
when the semi-annual inequality is eliminated, (by
taking a mean of the two half-years so that their oppo-
site deflections may counteract each other,) the residual,
or mean diurnal variation is extremely small, and is such
as corresponds to a station very little removed from the
line where it would vanish altogether ; whilst at Algiers,
the mean diurnal variation has a range of about 7/05
(Aime, p. 19), which is little less than at Toronto or
Hobarton. Thus, in this respect also, Algiers has the

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 509

character of a middle latitude station, and the Cape, that
of an equatorial station.

The facts thus cited seem to be quite irreconcilable
to the thermic or thermo-electric- current hypotheses ;
and not less so to M. Arago's belief that the diurnal
variation would be found to have changed completely
" du moment ou le soleil a passe d'un cote du zenith a
1'autre : " nor do they sanction the opinion that the line
of no dip should be regarded as the magnetic equator.

It seems to follow from what has been said, that a
physical explanation of the facts of the semi-annual
inequality will have to account for the epochs at which
the half-yearly phases pass into each other, which
epochs appear to respect the sun's passage of the geo-
graphical equator, — and also for the terrestrial distri-
bution of the phenomena in question, which appears to
respect the magnetical equator, or the line of least
magnetic force.

It has been stated above, that the turning hours, or
hours of greatest easterly and greatest westerly deflection
of the semi-annual portion of the diurnal variation, are
from 7 to 9 A.M., and from noon to 2 P.M. On pursuing
this correct general statement into still more precise
details, we find, within the above-named limits, a re-
markable difference between the two half-years, strik-
ingly analogous to a similar difference pointed out in
the preceding section, as existing between the exact
turning hours of the regular solar-diurnal variation in
the two hemispheres. When the sun is in the northern
signs, and the direction of the movement pertaining to
the semi-annual inequality agrees with that of the
regular solar-diurnal movement throughout the year in

510

the northern magnetic hemisphere, the times of the
semi-annual inequality movement are systematically
earlier than in the opposite half-year, when the sun is
in the southern signs, and when the direction agrees
with that of the regular solar-diurnal movement of the
southern hemisphere : just as the times of the regular
solar-diurnal movement are throughout the year earlier
in the northern than in the southern hemisphere. In
this minor detail, as well as in the broader features, we
have similar semi-annual and hemispherical agreements
and differences. The distinctive character of the two
half-years is seen in its simplest form at magnetically-
equatorial stations, where the regular solar-diurnal varia-
tion drops out altogether, and the semi-annual inequality
constitutes the sole diurnal variation (during the hours
of the day). It is well shown (in the forenoon extremes
particularly) in the Phil. Trans. 1847, Art. VI. Plates
3 and 4, where the analogy with the hemispherical dif-
ference of the regular solar-diurnal variation is also
pointed out. It seems, as far as can be yet judged, to
be a general and systematic natural feature, and one too
important to be safely overlooked by those who are
seeking for physical explanations of the minor magnetic
variations.

§ 3. — Diurnal Variation caused by the Disturbances.

IT has been stated in Note II., that amongst the perio-
dical laws by which the mean effects of the disturbances
are governed, there is one which has for its period a
mean solar day. We have still, therefore, to consider

4

SOLAR-DIURNAL VARIATION OP THE DECLINATION. 511

the additional complication which is occasioned by the
superposition of this disturbance-diurnal variation upon
the variations treated of in § 1 and § 2. Respecting
the disturbance-variation generally, it must be noticed
in the first place that, although it presents at each of
the stations where it has yet been examined, the same
characteristics of a regular, systematic, and easily deter-
minable progression, with well-marked hours of maxima
and minima, these hours, unlike the turning hours of
the variations treated of in the two preceding sections,
vary at different stations apparently without limit. It
had been at one time inferred from observations at a
European station, ee that the disturbance- variation has
always the same direction as the regular solar-diurnal
variation, and thus the disturbing forces tend only to
increase the action of the forces which produce the
solar-diurnal variation ; " this conclusion, perfectly well-
founded as regards the evidence derivable from one
station, may be met with equal truth by an exactly
opposite one at another station. For example, at two
places on the American continent, Toronto and Point
Barrow, the regular solar-diurnal variation has the same
turning hours, and the same direction at both stations.
But the disturbance-variation turns, indeed, at both
stations, nearly at the same hours, but does so in oppo-
site directions, the greatest westerly extreme of the
disturbance-variation at the one station being reached
nearly at the same hour as the greatest easterly at the
other station, and vice versa ; so that at one station the
two variations (as in the European case above referred to),
agree with and reinforce each other ; while at the other
station they oppose, and, to a certain extent, counteract

512 EDITOR'S NOTES.

each other. There seems reason to believe that the general
characters of the regular solar-diurnal variation, in regard
to the hours of its easterly and westerly extremes, are
the same in all the extra-tropical portions of the same
magnetic hemisphere ; whilst in the case of the disturb-
ance-variation, it seems probable that every diversity
in the times of occurrence of the turning-hours may be
found in different localities. The stations at which
these laws have been elicited are not yet sufficiently
numerous and varied to afford sufficient indication of
the mutual connections, or of the geographical or mag-
netical relations, which may possibly hereafter be traced
and throw additional light on the nature of the causes
concerned. The easterly and westerly disturbance-
deflections at the same station have also distinct laws oi
diurnal variation, which appear to have no necessary
or determinate relation to each other, and are only to
be learnt, like their conjoint effects, by examination.
The determination of these effects with a view to their
elimination, is a first step towards a true presentation of
the phenomena of the regular diurnal^variation, inas-
much as prior to such determination we cannot antici-
pate, with any degree of certainty, at what hour of the
twenty-four either the easterly or the westerly extreme
disturbance-deflection may be found : nor, in reference to
the comparative magnitude of the disturbance-variation
and the variation from other sources upon which it is
superposed, can we with confidence affirm more than
that the disturbance-variation may be expected to be
smallest at magnetically equatorial stations, — quite suffi-
ciently large in the middle latitudes to exercise fre-
quently a very perplexing influence on results from

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 513

which it has not been eliminated, — and in still higher
latitudes (more particularly when we approach the
localities from whence the disturbances appear to ema-
nate), so great as to mask altogether the diurnal vari-
ation proceeding from other sources.

It has been stated that the principal deflections
caused by the diurnal variation in the course of the
twenty-four hours, are those which take place during
the hours of the day. We have now to consider the
minor phenomena that occur during the hours of
the night. The progressive march of the declination-
magnet from the extreme elongation which it reaches in
the early hours of the afternoon to its opposite extreme
reached in the forenoon hours of the following day, — (a
movement of the north end of the magnet from west to
east in the extra-tropical parts of the northern magnetic
hemisphere, and from east to west in the same parts of
the southern magnetic hemisphere,) — is found to be
interrupted in both hemispheres by a minor retrogres-
sion (of variable duration); after which the previous
direction is resumed. The phenomenon of this " noc-
turnal episode," as it has been not inappositely called,
appears to have been first observed and noticed by
M. de Humboldt himself, in conjunction with M. Gray
Lussac, at Eome, in 1805 (p. 127); but although a very
important feature, it seems to have been unaccountably
lost sight of, for it is wholly unnoticed byM. Arago in 1836
(Annuaire, p. 283), when stating in detail the diurnal
march of the declination in the northern and southern
hemispheres, On the reception at Woolwich of the
earliest observations of the British colonial observato-
ries, it was at once perceived, from the examination of

IV. L L

514 EDITOR'S NOTES.

the records at Toronto and Hobarton, that the noc-
turnal retrogression is a persistent feature in both
hemispheres, — and it was at the same time recognised
that the hours at which this interruption of the other-
wise apparently regular progression took place, were
those which were most affected by the disturbances;
and that (in both hemispheres), the interruption was
precisely such as might be occasioned by them, and
might consequently be made to disappear if their
influence could be wholly eliminated. The following
extract from the first volume of the "Toronto Obser-
vations," published in 1845, will show that such was the
view taken even at that early period. " The influence
of the disturbances must now be regarded as a regular
component part of the diurnal variation itself (mean
quantities being considered). They now, therefore,
pass from the domain of that portion of the observed
phenomena to which we give the name of accidental or
irregular, (from our present ignorance of the laws which
regulate them), to be classed with the other portion in
which we recognise laws of order and succession, though
as yet we may not be able to assign either causes or
precise numerical values. On the one hand we cannot
with propriety overlook their influence on the diurnal
variation, because though they are of only occasional
occurrence, they yet form a permanent and regular
part of its mean value ; diminishing (in mean values)
the easterly direction of the north end of the mag-
net in the forenoon, and increasing it in the evening.
On the other hand, it becomes an object of much
interest to learn what portion of the diurnal variation is
thus due to what is apparently an effect of a distinct

SOLAR-DIURNAL VARIATION OF THE DECLINATION. 515

and special cause, and how the residual portion of the
variation would be affected by its separation ; and this
interest is increased when, even on a slight examination,
we perceive that the hours which are chiefly affected by
disturbance are those hours of the night when the con-
tinuous, and otherwise regular, easterly progression of
the diurnal variation suffers interruption, occasioning a
double, and even sometimes in the winter months a
triple, alternation of easterly and westerly movement."
(Tor. Obs. vol. i. p. xxi.)

The view thus early taken has been strengthened by
the results of the observations in succeeding years. The
third volume of the same work contains a discussion of
the diurnal variation obtained from five years of hourly
observation at Toronto, wherein it is shown (pp. Ixxxvi
— Ixxxix) that when all the disturbances are allowed to
remain in the body of the observations, the nocturnal
westerly retrogression commences a little after 9 P.M.,
and the easterly point which the declination had reached
at that hour is not regained until after 4 A.M., having
been at 2 P.M. 0'.7 west of that point : but that, when the
larger disturbances, or those which equalled or exceeded
Z on either side of the mean or normal at the same hour
in the same month, are eliminated, the westerly retrogres-
sion is diminished in duration to the interval between 1 1
P.M. and 3 A.M., and in value to 0''19 at 2 P.M. ; and fur-
ther, that if we proceed on the assumption that by ab-
stracting the disturbances equalling or exceeding 5', we
may have eliminated half of the whole influence of the
causes to which the larger disturbances are due (which
is certainly no unreasonable supposition), the westerly
retrogression becomes almost wholly obliterated, and the

516 EDITOR'S NOTES.

residual diurnal variation is virtually a single progression
with but one maximum and one minimum.

The belief that the " nocturnal episode " is, in great
part, at least, if not wholly, the effect of the disturbances,
is further strengthened by the analogous facts of the
diurnal variation of the total magnetic force at Toronto,
discussed in pp. xciii — xcv of the same volume. When
the diurnal variation of this element is obtained from
the whole of the observations, including the disturb-
ances, two well marked maxima and minima appear in
the twenty-four hours ; but by eliminating the disturb-
ances in a greater or less degree, by a process similar to
that adopted in the declination, one of the two maxima
and one of the two minima diminish in proportion to
the degree in which the disturbances are removed,
giving full reason to believe that by eliminating their
influence altogether, the residual diurnal variation would
be reduced to the simple form of a single progression in
twenty-four hours.

There is also reason to believe that the supposed ano-
malies in the turning hours of the diurnal-variation which
have been observed in the high latitudes, and which are
adverted to in pages 129 and 130 of this volume, are pro-
duced by the disturbances. At the only station in the
high latitudes where the large disturbances have been se-
parated, viz. Point Barrow (Phil. Trans. 1857, Art xxiv.),
the effect of the separation has been to remove the mask
by which the true features of fche solar-diurnal variation
were concealed, and to restore its turning hours
accordance with the order of the phenomena observ
generally in the same hemisphere.

:
rf

NOTES.

(*) p. 14. — Kosmos, Bd. iii. S. 107 (English edition, p. xv. Note 94);
compare also Bd. ii. S. 464 and 508 (English edition, p. Ixxviii. and cxiii.).

(2) p. 18. — " La loi de 1'attraction reciproque au carre' de la distance est celle
des emanations qui partent d'un centre. Elle parait £tre la loi de toutes les
forces dont 1'action se fait apercevoir a des distances sensibles ; comme on 1'a
reconnu dans les forces electriques et magnetiques. Une des proprie'tes re-
marquables de cette loi est que, si les dimensions de tous les corps de 1'univers,
leurs distances mutuelles et leurs vitesses venaient a croltre ou a diminuer
proportionnellement, ils de'criraient des courbes entierement semblables a celles
qu'ils decrivent: en sorte que 1'univers, reVluit ainsi successivement jusqu'au plus
petit espace imaginable, offrirait toujours les me'mes apparences aux observateurs.
Ces apparences sont par consequent inde'pendantes des dimensions de 1'univers,
comme, en vertu de la loi de la proportionalite* de la force a la vitesse, elles sont
inde'pendantes du mouvement absolu qu'il peut y avoir dans 1'espace." Laplace,
Exposition du Syst. du Monde (5me e'd.), p. 385.

(3) p. 19. — Gauss, Bestimmung des Breitenunterschiedes zwischen den
Sternwarten von Gb'ttingen und Altona, 1828, S. 73. By a singular accident,
the two observatories are, within less than the width of a house, in the same
meridian.

(4) p. 19. — Bessel, fiber den Einfluss der Unregelmassigkeiten der Figur der
Erde auf geodatische Arbeiten und ihre Vergleichung mit astronomischen Be-
stimmungen, in Schumacher's Astron. Nachr., Bd. xiv. No. 329, S. 270 ; and
also Bessel and Baeyer, Gradmessung in Ostpreussen, 1838, S. 427 — 442.

(5) p. 20. — Bessel, liber den Einfluss der Veranderungen des Erdkorpers auf
die Polhb'hen, in Lindenau and Bohnenberger, Zeitschrift fur Astronomic, Bd. v-

VOL. IV. a

11 NOTES.

1818, S. 29.' " The weight of the Earth expressed in pounds = 9933 x 1021,
and the supposed displaced mass = 947 x 101V

(6) p. 20. — The theoretical investigations of that period were followed by
those of Maclaurin, Clairaut, D'Alembert, Legendre, and Laplace; to tfiese
should be added the theorem propounded in 1834 by Jacobi: that ellipsoids
having three unequal axes may, under certain conditions, be as good " figures
of equilibrium as the two earlier assigned ellipsoids of revolution." See, in Pog-
gendorffs Annalen der Physik und Chemie, Bd. xxxiiL 1834, S. 229 — 233,
the memoir of Jacobi, removed by an early death from his friends and ad-
mirers.

(7) p. 21. — The first exact comparison of a considerable number of measure-
ments of arcs (i. e. the Quito, two Indian, French, English, and recent
Lapland arcs) was undertaken, with much success, in 1819, by Walbeck at
Abo. He found the mean values of —^ for the ellipticity, and 57009*,? 58
for the length of a mean degree of the meridian. This memoir, entitled " De
Forma et Magnitudine Telluris," has not, unfortunately, been published entire. At
the recommendation of Gauss, Eduard Schmidt, in his valuable " Lehrbuch" of
mathematical geography, has repeated the investigation with improvements,
taking into account the higher powers of the ellipticity as well as the latitudes
observed at intermediate points ; and adding the Hanoverian arc, and the prolon-
gation by Biot and Arago of the French arc to Formentera. The results, made
successively more perfect, were published in three forms : in Gauss's " Bestim-
mung der Breitenunterschiede von Gottingen und Altona," 1828, S. 82; in
Eduard Schmidt's " Lehrbuch der mathem. und phys. Geographic," 1829, Th. i.
S. 183 and 194—199; and lastly, in the Preface to the same book, S. v. The
final result is: Degree of the meridian, 57008t>655; ellipticity, ~^. Bessel's
first investigation was immediately preceded (1830) by Airy's important Me-
moir on the Figure of the Earth in the Encyclopaedia Metropolitana, pp. 220 and
239, in the edition of 1849 (Semi-polar axis, 20853810 feet = 32611637
toises ; semi-equatorial axis, 20923713 feet = 3272095'2 toises ; meridian
quadrant, 32811980 feet = 5131208-0 toises; ellipticity, ~^)- Our
great Konigsberg astronomer, Bessel, was constantly occupied from 1836 to
1842 with calculations on the figure of the Earth, and as his earlier investiga-
tions were amended and improved by later ones, the mixture of results obtained
at different times has become a fertile source of error in many writings. In the
case of numbers, which are, by their nature, mutually dependent on each other,
this kind of confusion, sometimes made worse by incorrect reduction of measures,
(toises, metres, English feet, and miles 60 or 69 to an equatorial degree), is the
more to be regretted as it causes enquiries which have cost much labour and

NOTES. Ill

time to appear in a disadvantageous light. In the summer of 1837, Bessel
published two memoirs ; one on the influence of the irregularities of the Earth's
figure on geodesical measurements, and their comparison with astronomical de-
terminations ; and the other on the axes of the elliptic spheroid of rotation which
would correspond best with the existing measurements of arcs of meridians.
(Schum. Astr. Nachr., Bd. xiv. No. 329, S. 269, and No. 333, S. 345.) The
results of the calculation were: — Semi-major axis, 3271953*'854 ; semi-minor
axis, 3261072t-900; length of a mean degree of the meridian, i. e. the ninetieth
part of a quadrant of the Earth, in the direction perpendicular to the equator,
57011tg453. An error of 68 toises, discovered by Puissant, in the mode of
calculation employed by a commission of the French National Institut, 1808,
in determining the distance of the parallels of Monljouy, near Barcelona,
and Mola, in Formentera, occasioned Bessel to subject his first work on the
dimensions of the Earth to a fresh revision, in 1841. (Schum. Astr. Nachr.,
Bd. six. No. 438, S. 97—1 1 6.) The revision gave for the length of the Earth's
quadrant 513 11 79f-81, — (5130740 toises had been supposed in the first de-
termination of the metre), — and for the mean length of a degree of the meridian
57013{-109 (being 0*-611 more than the degree of the meridian in 45° lat.).
The numbers given in the text are the results of this, Bessel's last determina-
tion. The length of 5131180 toises as the quadrant of the meridian (with a
mean error of 255t<63), is equivalent to 10000856 metres; making the entire
circumference of the Earth 40003423 metres (or 5390'98 German, or 21563*92
English geographical miles). The difference from the original assumption of
the Commission des Poids et Mesures, according to which the metre should be
the 40-millioneth part of the Earth's circumference, amounts, therefore, for the
circumference, to 3423met, or 1756t-27, i. e. 0'46 of a German, or 1'84 of an En-
glish geographical mile. According to the earliest determinations, the length of
the metre was fixed atOt'5130740; according to Bessel's latest determination, it
should be O'-SISIISO. The difference for the length of the metre is therefore
0-038 Paris lines, making it, according to Bessel, 443334, instead of 443296
Paris lines, which is its present legal value. Compare also on this so-called
" natural standard," Faye, Le9ons de Cosmographie, 1852, p. 93.

(8) p. 23.— Airy, Figure of the Earth, in the Encycl. Metrop. 1849, p. 214 —
216.

(9) p. 23.— Biot, Astr. physique, t. ii. p. 482 and t. iii. p. 482. A very ex-
act measurement of a parallel arc has been made by Corabceuf, Delcros, and
Peytier, on the " Parallel of the Chain of the Pyrenees ;" a work of the greater
importance because it has led to the comparison of the levels of the Mediterra-
nean and the Atlantic.

a 2

IV NOTES.

(10) p. 24. — Kosmos, Bd. i. S. 175 (English edition, p. 158). " II est tres-
remarquable qu'un astronome, sans sortir de son observatoire, en comparant
seulement ses observations a 1'analyse, eut pu de'terminer exactement la gran-
deur et I'aplatissement de la terre, et sa distance au soleil et a la lune, e'le'mens
dont la connaissance a e'te' le fruit de longs et pe*nibles voyages dans les deux
hemispheres. Ainsi la lune, par 1'observation de ses mouvements, rend sensible
a 1'astronomie perfectionne'e Tellipticite de la terre, dont elle fit connaftre la
rondeur aux premiers astronomes par ses ellipses." (Laplace, Expos, du Syst.
du Monde, p. 230.) We have already noticed, in Bd. iii. S. 498 and 540 (En-
glish edition, p. 355 and cxxvii), an almost analogous optical proposal of Arago,
founded on the remark that the intensity of the ash-coloured light (i. e. of the
" earthlight" on the moon) might inform us as to the mean state of transparency
of the whole of our atmosphere. Compare also Airy, in the Encycl. Metrop. p.
189 and 236, on the determination of the Earth's ellipticity from the moon's
movements, and p. 231—235, on deductions of the Earth's ellipticity from the
observed magnitudes of precession and nutation. According to Biot, procession
and nutation could only give for the ellipticity limitary values (^ and 3^), very
distant from each other. (Astron. physique, 3e e'd. t. ii. 1844, p. 463.)

(") p. 24. — Laplace, Me'canique celeste, e'd. de 1846, t. v. p. 16 and 53.

(12) p. 24.— Kosmos, Bd. i. S. 421. Anm. 1 (English edition, p. xliii. Note
131). The application of the isochronism of the vibrations of a pendulum in
Arabian astronomical writings was first made known by Edward Bernard in
England. See his letter from Oxford, April, 1683, to Dr. Robert Huntington, in
Dublin. (Phil. Trans, vol. xii. p. 567.)

(13) p. 24. — FreVet, de 1'^ltude de la Philosophic ancienne, in the Mem. de
1'Acad. des Inscr. t. xviii. (1753) p. 100.

(14) p. 25.— Picard, Mesure de la Terre, 1671, art. 4. It is hardly probable
that the conjecture expressed at the Paris Academy before 1671, of a variation
of the intensity of gravity with the latitude (Lalande, Astronornie, t. iii. p. 20, §
2668), should have belonged to the great/ Huygens. Huygens had indeed pre-
sented his " Discours sur la cause de la Gravitd" to the Academy in 1669 ; it is
not, however, in this memoir, but in the " additamentis," one of which must have
been completed after the appearance of Newton's Principia, as it refers to it —
therefore, after 1687 — that Huygens speaks of the shortening of the second's pen-
dulum observed by Richer at Cayenne. He says: "Maxima pars hujus libelli
scripta est, cum Lutetian degerem(to 1681), ad cum usque locum, ubi de altera-
tione, qua3 pendulis accidit e motu Terra;." Compare the explanation given by
me in Kosmos, Bd. ii. S. 520, Anm. 2 (English edition, p. cxxv. Note 542).
The observations made by Richer at Cayenne were not published until fully six

NOTES. V

years after his return, and, what is more surprising, no notice of his important
double observation, of the clock pendulum and of a simple second's pendulum,
occurs in the Proceedings of the Acade'mie des Inscriptions during this long in-
terval. We do not know when Newton, whose earliest theoretical speculations
on the figure of the Earth were antecedent to 1665, first became aware of
Eicher's results. Picard's arc measurement, although published in 1671, only
came very late (1682), and indeed accidentally, in conversation at a meeting of
the Royal Society, to Newton's knowledge. It exercised, as Sir David Brewster
has shown (Life of Newton, p. 152), an exceedingly important influence on his
determination of the Earth's diameter and the relation of the fall of bodies on our
planet to the force which governs the moon in her course. We may suppose a
similar influence to have been exercised on Newton's ideas by the knowledge of
the elliptic figure of Jupiter, which was recognised by Cassini previous to 1666,
but first described by him in 1691, in the Me'moires de 1' Academic des Sciences,
t. ii. p. 108. May Newton have known anything of an earlier publication, some
sheets of which were seen by Lalande in Maraldi's hands ? (Compare Lalande,
Astr. t. iii. p. 335, § 3345, with Brewster's Life of Newton, p. 162, and Kos-
mos, Bd. i. S. 420, Anm. 99; English edition, p. xlii. Note 129.) In the con-
temporaneous labours of Newton, Huygens, Picard, and Cassini, the great delay
in publication which was then customary, and the frequent accidental delays in
communications, render it very difficult to arrive at any well-assured traces of
the transmittance or exchange of scientific ideas.

(15) p. 26.— Delambre, Base du Syst. me'trique, t. iii. p. 548.

(16) p. 26.— -Kosmos, Bd. i. S. 422, Anm. 3 (English edition, p. xlv. Note
133); Plana, Ope'rations geode'siques et astronomiques pour la Mesure d'un Arc
du Parallele moyen, t. ii. p. 847 ; Carlini, in the EfFemeridi astronomiche di
Milano per 1'anno 1842, p. 57.

(17) p. 26. — Compare Biot, Astronomic physique, t. ii. (1844) p. 464, with
Kosmos, Bd. i. S. 424, end of Anm. 3 (English edition, Note 133), and Bd. iii.
S. 432 (English edition, p. 310), where I notice the difficulties which the com-
parison of the planet's time of rotation with its observed ellipticity presents.
Schubert (Astron. Th. iii. S. 316) had called attention to this difficulty. Bessel,
in his memoir " Ueber Haass und Gewicht," says expressly that " the assump-
tion of gravity remaining always the same at a place where it has been observed,
appears in some degree uncertain, from the knowledge we have acquired of the
slow changes of level of considerable portions of the Earth's surface."

(18) p. 26. — Airy, in his excellent Memoir on the Figure of the Earth (Encycl.
Metrop. 1849, p. 229), counted, in the year 1830, fifty different stations at which
well-assured results had been obtained ; and fourteen others where the results

VI NOTES.

(obtained at earlier periods by Bouguer, Legentil, Lacaille, Maupertius, and La
Croyere) could not be looked upon as having nearly the same degree of exact-
ness.

(19) p. 28. — Biot and Arago, Recueil d'Observ. ge'ode'siques et astronomiques,
1821, p. 526—540, and Biot, Traitd d'Astr. phys. t. ii. 1844, p. 465—473.

(20) p. 28. — Idem, p. 488. Sabine (Exper. for determining the Variation in
the Length of the Pendulum vibrating Seconds, 1825, p. 352) found from his
thirteen stations, so widely dispersed over the northern hemisphere, ^5 ; and
from the same with the addition of all the pendulum stations of the British Sur-
vey and of the French arc of the meridian (from Formentera to Dunkirk) — from
the comparison therefore, in all, of 25 points of observation — almost the same re-
sult, viz. ~^. At a considerable distance from the above stations (situated
on the two sides of the Atlantic), the pendulum lengths found in the meridians of
Petropaulowski and Sitka, give, as Liitke has remarked, a still much higher
ellipticity : 5^. Bessel, with his peculiar lucidity, has shown analytically, in
his investigations on the length of the simple second's pendulum (S. 32, 63, and
126 — 129), how the theory of the influence of the air surrounding the pendulum
which was previously generally received and employed, leads to an error in cal-
culation, and renders a correction necessary, on account of the difference of the loss
of weight of solid bodies when they are immersed in a fluid, either in a state of
repose or in vibration. Such a correction was somewhat obscurely pointed to in
1 786, by the Chevalier de Buat, Bessel says, " If a body moves in a fluid (air),
the latter becomes part of the moving system ; and the moving force must be im-
parted, not only to the particles of mass of the solid body put in motion, but also
to all the particles of mass of the fluid which are put in motion." On the experi-
ments of Sabine and Baily, to which Bessel's practically important pendulum
correction (reduction to a vacuum) gave occasion, see John Herschel in his
Memoir of Francis Baily, 1845, p. 17 — 21.

(21) p. 28.— Kosmos, Bd. i. S. 175 and 422, Anm. 2 (English edition,
p. 158 ; Note 132). Compare for the phenomena in islands, Sabine, Pendulum
Exper. 1825, p. 237 ; and Lutke, Obs. du Pendule invariable, exe'cute'es de
1826 — 1829, p. 241. The same work contains a remarkable table of the kinds
of rocks at sixteen pendulum stations (p. 239) from Spitsbergen (79° 50' N.
lat.) to Valparaiso (in 33° 2' S. lat.). [See also Sabine, Pend. Exper. 1825,
p. 338 ; in which is the first table of this description, containing thirteen sta-
tions.—Ed.]

(K) p. 29. — Kosmos, Bd. i. S. 424, Anm. 5 (English edition, Note
135). Eduard Schmidt (Mathem. und phys. Geographic, Th. i. S. 394)
has taken from the many pendulum observations made in the corvettes Descu-

NOTES. Vll

Uerta and Atrevida, under the command of Malaspina, thirteen stations be-
longing to the southern hemisphere, and found, on the mean, an ellipticity
of 2t^.54. Mathieu also derived from Lacaille's observations at the Cape of
Good Hope, and in the Isle of France, compared with Paris, ~-4 ; but the
measuring apparatus of that period did not offer the security afforded by that of
Borda and Kater, and the later methods of observation. This is the place for
mentioning the fine experiments of Foucault, so honourable to the sagacity of
the inventor, which have given a visible proof of the rotation of the Earth on
its axis by means of the pendulum, its plane of vibration turning slowly ffom east
to west. (Comptes rendus de 1'Acad. des Sciences, se'ance du 3 fe'vrier, 1851,
t. xxxii. p. 135.) To obtain deviations to the eastward, as in the experi-
ments on falling bodies of Benzenberg and Reich on steeples and in shafts, a
very considerable height of fall is required ; whereas Foucault's apparatus
renders the rotation of the Earth sensible with a length of pendulum of six feet.
Phenomena which are explained by the rotation of the Earth (as the march of
Richer's clock at Cayenne, diurnal aberration, deflection of projectiles, and
trade winds), are not to be placed in comparison with that which can at all
times be called forth by Foucault's apparatus, and of which the members of the
Academia del Cimento seem to have had a conception, but not to have pursued
it. (Antinori, in the Comptes rendus, t. xxxii. p. 635.)

(23) p. 30. — In Grecian antiquity, two parts of the Earth were pointed out as
having, according to inferences drawn from the opinions which then prevailed,
remarkable inflation or elevation of surface : the north of Asia and the country
under the equator. " The high and naked Scythian plains," said Hippo-
crates (De Mice et Aquis, § xix. p. 72, Littre), " without being crowned with
mountains, spread out and rise to beneath the Bear." The same belief had been
attributed to Empedocles at a yet earlier epoch (Plut. de Plac. Philos. ii. 8).
Aristotle (Meteor, i. 1 a 15, p. 66, Ideler) says, that the older meteorologists,
who "made the sun pass, not under, but round the Earth, considered the swelling
of the Earth towards the north to be the cause of the nocturnal disappearance
of the sun, and thus of the occurrence of night." Also, in the compilation of the
problems (xxvi. 15 ; Pag. 941, Bekker), the coldness of the north wind is
attributed to the " height of the ground " in that quarter. In all these passages
it is not mountains, but high plains, or general elevation of the land, which are
referred to. I have already shown elsewhere (Asie Centrale, t. i. p. 58), that
Strabo, — who is the only writer who employs the characteristic word opoir&ia
for Armenia (xi. p. 522, Casaub.), for Lycaonia, inhabited by the wild ass
(xii. p. 568), and for Upper India, and the gold country of the Derdans (xv.
p. 706), — always distinguishes the difference of climate due to geographical

Vlll NOTES.

latitude from that which belongs to difference of elevation. " Even in southern
regions," says the geographer of Amasia, " every high ground, even if it is a
plain, is cold." (ii. p. 73.) Eratosthenes and Polybius adduce in favour of a
very moderate temperature at the equator, not only the more rapid passage of
the sun (Geminus, Elem. Astron. c. 13 ; Cleom. Cycl. Theor. i. 6), but also,
and more especially, the swelling or elevation of the ground. (See my Examen
crit. de la Ge'ogr. t. iii. p. 150 — 152.) Both those writers, according to the
testimony of Strabo (ii. p. 97), assert, that " the part of the Earth which is
near the equator is the highest, and that for this reason it is also very rainy,
because the winds, which change with the seasons, bring from the north a
great amount of clouds, which hang upon the high land." Of these two beliefs
of two great elevations of the ground, one in Northern Asia (the Scythian
Europe of Herodotus), and the other in the equatorial zone, the first has main-
tained itself, with that pertinacity which belongs to error, for almost two thou-
sand years, and has led to the geological fable of an uninterrupted high land of
Tartary north of the Himalaya ; while the second could only be supported for
an extra tropical part of Asia, for the "high" or " mountain "-plain of Meru,
or Mount Meru, celebrated in the oldest and noblest monuments of Indian
poetry. (See Wilson's Diet. Sanscrit and English, 1832, p. 674, where Meru
is interpreted " high plain.") I have thought it necessary to enter into these
details, in order to refute the hypothesis of the ingenious Fre'ret, who, without
quoting passages from Greek authors, and only alluding to a single one re-
lating to tropical rains, interprets the belief in local elevations of the ground, as
having reference either to a flattening of the Earth at the poles, or to the con-
verse. " Pour expliquer les pluyes," said Fre'ret (Mem. de 1'Acad. des Inscrip-
tions, t. xviii. 1753, p. 112), "dans les re'gions e'quinoxiales que les conquetes
d'Alexandre firent connoitre, on imagina des courans qui poussoient les nuages
des poles vers 1'e'quateur, ou, au de'faut des montagnes qui les arretoient, les
nuages 1'etoient par la hauteur ge'ne'rale de la Terre, dont la surface sons 1'e'qua-
teur se trouvoit plus e'loignee du centre que sous les poles. Quelques physicians
donnerent au globe la figure d'un sphe'roide renfle' sous 1'e'quateur et aplati vers
les poles. Au contraire, dans 1'opinion de ceux des anciens qui croyoient la Terre
alonge'e aux poles, le pays voisin des poles se trouvoit plus e'loigne' du centre que
sous 1'e'quateur." I cannot find any evidence in antiquity to justify these state-
ments. In the third section of the first book of Strabo (Pag. 48, Casaub.), it
is said expressly, " The whole Earth, as Eratosthenes has said, is globular, but
not as if turned by a turner " (an expression borrowed from Herodotus, iv. 36),
" having many irregular deviations from so exact a form, occasioned by water,
fire, earthquakes, subterranean gusts of wind (elastic vapours ?), and other such

NOTES. IX

causes. The globular form of the entire earth follows from the arrangement of
the whole, which is not altered by such minor disfigurements ; the little disap-
pears in the great." Subsequently it is said, following always Groskurd's very
successful version, that " the Earth, with the sea, is globular, the land and seas
forming but one surface ;" that " the rising of the land out of the water being
inconsiderable, in comparison with the whole extent, must remain unnoticed.
We do not suppose the form of the globe to be as if turned on a turning-lathe,
or conforming to the exact measurement which an artist might make of an
artificial ball ; but rather speak of its roundness as we might of the roundness
of such a ball, judged of more roughly by the eye." (Strabo, ii. p. 112.) " The
world is the work both of Nature and Providence : of Nature, inasmuch as all
tends together towards a point in the middle of the whole, around which it
arranges itself the less dense (water), outside that which is more dense (earth)."
(Strabo, xvii. p. 809.) When the form of the Earth is spoken of among the Greeks
(Cleom. Cycl. Theor. i. 8, p. 51), it is compared simply to a disc; flat, or having
a depression in the middle; to a cylinder (Anaximander) ; to a cube; or to a
pyramid ; and, finally (notwithstanding the long controversy of the Epicureans,
who denied the attraction to the centre), it came to be generally regarded as a
globe. The idea of the flattening at the poles, which we call compression or
ellipticity, did not, however, present itself. The " longish " Earth spoken of
by Democritus is only the " disc " of Thales, longer in one direction than in the
other. The TO (rx^/w« rvfjnravoeities, an idea attributed particularly to Leu-
cippus (Plut. de Plac. Philos. iii. 10; Galen. Hist. Phil. cap. 21 ; Aristotl. de
Co3lo, ii. 13 ; Pag. 293, Bekker), is at the bottom of the representation of a
hemisphere, or half-globe, with an even base, which, perhaps, designates the
equator, while the curvature is regarded as the oiKov/j-evrj. A passage in Pliny
(ix. 54), on pearls, illustrates this form ; while, on the other hand, Aristotle,
Meteorol. ii. 5 a 10 (Ideler, t. i. p. 563), only offers a comparison of segments
of a globe with the tympanum, as also appears from the Commentary of Olym-
piodorus (Ideler, t. i. p. 301). I have purposely not noticed in this review two
passages very familiar to me (Agathemer, de Geographia, lib. i. cap. i. p. 2,
Hudson); Eusebius (Evangel. Prseparat. t. iv. p. 125, ed. Gaisford, 1843), be-
cause they show how inaccurately subsequent writers have often attributed to
the ancients opinions which were, in truth, entirely strange to them. According
to these passages, Eudoxus would have spoken of the Earth as a disc, having a
length twice as great as its breadth ; as would also Dicearchus, the disciple of
Aristotle, although he had actually proposed proofs of its globular form (Mar-
tian. Capella, lib. vi. p. 192) ; while Hipparchus would have regarded the Earth
as Tpa7re£oei57js, and Thales would have held it to be spherical !

X . NOTES.

(24) p. 30.— Bessel, in a letter to myself, Dec. 1828, says : " It often appears
to me as if the Earth's ellipticity was sometimes spoken of as uncertain, because
too great exactness is demanded. If we take gfo, 3^, ^ and 3^, the corre-
sponding differences between the greater and the lesser radius are 10554,
10905, 11281, and 11684 toises. The change of 30 units in the denominator
only alters 1130 toises in the polar radius; an amount which, viewed in
relation to the visible inequalities of the Earth's surface, appears so immaterial,
that I am really often astonished that the experiments should agree within such
near limits. Detached observations, isolated in wide spaces, would indeed teach
us little more than we know at present ; but it would be important to unite all
the measurements over the entire surface of Europe, bringing all the astro-
nomically determined points into the operation." Thjs, however, would still only
show rtie form of the Earth in what may be termed the western peninsula, or
projecting part of the great Asiatic continent, about 66° of longitude. The
steppes of Northern Asia, even the central Kirghis Steppe, of which I have seen
a considerable portion, are often hilly, and, on the whole, not to be compared in
point of horizontality with the Pampas of Buenos Ayres and the Llanos of Vene-
zuela. These South American plains being distant from the mountains, and
covered with secondary and tertiary strata of very uniform and moderate density,
the anomalies which might be presented by pendulum experiments at different
parts of their surface would afford very pure, and, therefore, decisive results, as
to the local constitution of the deep interior terrestrial strata. Compare my
Ansichten der Natur, Bd. i. S. 4, 12, and 47—50.

(25) p. 31. — Bouguer, who invited La Condamine to undertake experiments
on the deflection of the plumb line by the Chimborazo, does not indeed mention
Newton's proposal in his " Figure de la Terre," p. 364 — 394. Unfortunately,
the best informed of the two travellers did not observe at opposite sides (east and
west) of the great mountain, but (Dec. 1738) at two stations on the same side,
one being south 6l£° west (distant 4572 toises from the centre of the mountain-
mass), and the other south 16° west (distant 1753 toises). The first was in
a part well known to me, probably below the height where the small mountain-
lake of Yana-Cocha is situated, and the other in the pumice -covered plain of the
Arenal. (La Condamine, Voyage & 1'Equateur, p. 68 — 70.) Contrary to all
expectation, the deflection indicated by the star altitudes was only 7"-5; which
was attributed by the observers themselves to the difficulties attendant on ob-
servations made so near the limits of perpetual snow, the inexactness of the in-
struments, and, above all, to supposed great cavities in the colossal trachytic
mountain. I have expressed, on geological grounds, considerable doubts as to the
justness of the assumption of these great cavities and consequent small mass of

NOTES. XI

the trachytic dome of Chimborazo. To the south-south-east of Chimborazo,
near the Indian village Calpi, is situated the eruption- cone of Yana-Urcu, which
I examined carefully with Bonpland, and which is certainly of more recent
origin than the elevation of its giant dome-shaped neighbour. We found nothing
resembling a crater in the latter. See the ascent of Chimborazo in my " Kleinere
Schriften," Bd. i. S. 138.

(M) p. 31. — Baily, Esper. with the Torsion Rod, for determining the mean
Density of the Earth, 1843, p. 6; John Herschel, Memoir of Francis Baily, 1845,
p. 24.

(") p. 32. — Reich, neue Versuche mit der Drehwage (New Experiments with
the Torsion Balance), in the Abhandl. der mathem. physischen Classe der Kb'n.-
Sachsischen Gesellschaft der Wissenschaften zu Leipzig, 1852, Bd. i. S. 405
and 418. The latest experiments of my excellent friend Reich are rather nearer
the fine investigation of Baily. I have derived the mean (5*5772) from the
series of experiments a and b : a. with the tin sphere and the longer and thicker
copper wire, 5*5712, with a probable error of 0'0113; b. with the tin sphere, and
the shorter and thinner copper wire, as well as with the tin sphere and the bifilar
iron wire, 5'5832, with a probable error of 0'0149. Taking into account these
probable errors of a and 6, the mean is 5'5756. Baily's result (5'660), obtained
indeed by more numerous experiments, might yet, however, give a rather too
high density, as its apparent value seemed to increase as the spheres employed
(glass or ivory) were lighter. (Reich, in Poggendorff's Annalen, Bd. Ixxxv. S.
190. Compare also Whitehead Hearn, in the Phil. Trans, for 1847, p. 217 —
229.) The motion of the torsion beam was observed by Baily according to the
method of Reich, by means of the image, reflected, as in the magnetic observa-
tions of Gauss, in a mirror attached to the middle of the beam. The highly
important use of the mirror, which so greatly increases the exactness of the
reading, was proposed so long ago as 1826. by Poggendorff (Annalen der Physik,
Bd. vii. S. 121).

(28) p. 33. — Laplace, Me'canique celeste, e'd. de 1846, t. v. p. 57. The mean
specific weight of granite is estimated at 27 at the most, as the specific weights
of the biaxal white mica, and the green uniaxal, are from 2'85 to 3*1, and those
of the rest of the constituents of the rock, quartz and felspar, are 2*56 and
2'65. Oligoclase itself is only 2'68, and although hornblende may be 3'17,
syenite, in which felspar always predominates, still remains much under 2'8.
As the weight of clay-slate is 2*69 — 278, as among limestones it is only in pure
dolomites that it attains 2'88, chalk 272, gypsum and rock salt 2'3, I consider
the density of the accessible continental crust of the Earth to be nearer 2'6 than
2'4. Laplace, on the assumption of the density increasing in arithmetical pro-

Xll NOTES.

gression from the surface to the centre, and with the certainly erroneous assump-
tion that the density of the outermost strata = 3, found, for the mean density
of the whole Earth, 4'7647, which differs considerably from the results of Reich,
5'577, and Baily, 5*660; far more so than the probable errors of observation could
at all account for. By a new discussion of Laplace's hypothesis, in an interest-
ing memoir which is soon to appear in Schumacher's Astr. Nachrichten, Plana
has arrived at the result that it is possible, by an alteration in the treatment
of this hypothesis, to represent very approximately Eeich's mean density of the
Earth, together with my estimate of 1/6 as the mean density of the outermost
stratum or surface of both land and sea, and the Earth's ellipticity, within the
probable limits of the last-named element. He says : " Si la compressibilite
des substances dont la Terre est forme'e a ete' la cause qui a donne' a ses couches
des formes re'gulieres, a peu pres elliptiques, avec une densite croissante depuis
la surface jusqu'au centre, il est permis de penser que ces couches, en se conso-
lidant, ont subi des modifications, a la verite fort petites, mais assez grandes
pour nous empecher de pouvoir driver, avec toute 1'exactitude que Ton pourrait
souhaiter, 1'e'tat de la Terre solide de son e'tat ante'rieur de fluidite. Cette re'-
flexion m'a fait appre'cier davantage la premiere hypothese, propose'e par 1'auteur
de la Mecanique celeste, et je me suis de'cide & la soumettre ^ une nouvelle dis-
cussion."

(a) p. 33. — Compare Petit, " Sur la latitude de 1'Observatoire de Toulouse, la
densite' moyenne de la chaine des Pyre'ne'es, et la probabilite' qu'il existe un vide
sous cette chaine," in the Comptes rendus de 1'Acad. des Sc. t. xxix. 1849, p.
730.

(M) p. 34.— Kosmos, Bd. i. S. 183 and 427, Anm. 10 (English edition,
p. 166, and Note 140).

(31) p. 34. — Hopkins (Physical Geology), Reports of the British Association
for 1838, p. 92; Phil. Trans. 1839, Pt. II. p. 381; and 1840, Pt. I. p. 193;
Henry Hennessey (Terrestrial Physics), in the Phil. Trans. 1851, Pt. II. p. 504
and 525.

(32) p> 34. —Kosmos, Bd. i. S. 249 and 450—452, Anm. 95 (English edi-
tion, p. 227, and Note 225).

(33) p. 35. — The observations communicated by Walferdin are of the autumn
1847. They differ very little from the results (Kosmos, Bd. i. S. 181, and
Anm. 8 (English edition, p. 163, and Note 138); Comptes rendus, t. xi. 1840,
p. 707), which Arago obtained also with Walferdin's apparatus in 1840, at a
depth of 505 metres, just as the borer had passed through the chalk and began
to penetrate the gault.

(**) p. 36. — According to manuscript communications from the mine captain

NOTES. Xlll

von Oeynhausen. Compare Kosmos, Bd. i. S. 416, Anm. 94, and S. 426, Anm. 8
(English edition, Notes 124 and 138) ; also Bischof, Lehrbuch der chem. und
phys. Geologie, Bd. i. Abth. i. S. 154 — 163. In absolute depth, the boring at
Mondorf, in the Grand Duchy of Luxembourg (2066 Fr. feet), comes next to
that of Neu-Salzwerk.

C5) p. 37. — Kosmos, Bd. i. S. 426 (English edition, p. 406), and Me'moires
de la Socie'te d'Hist, Naturelle de Geneve, t. vi. 1833, p. 243. The comparison
of a great number of Artesian wells in the neighbourhood of Lille with those of
St. Ouen and Geneva, might lead us to infer a more considerable influence of
the conducting power of the strata, if we could be satisfied that the data were
all equally accurate. (Poisson, The'orie mathe'matique de la Chaleur, p. 421.)

(*) p. 37. — In a table of 14 borings, of above 100 metres in depth, from the
most different parts of France, Bravais, in his instructive encyclopedic memoir,
" Patria," 1847, p. 145, cites nine in which the depth corresponding to an in-
crease of temperature of 1° centigrade falls between 27 and 39 metres, varying
from 5 to 6 metres on either side of the mean (32m) given in the text. (Com-
pare also Magnus in Poggend. Ann. Bd. xxii. 1831, S. 146.) On the whole,
the increase of temperature appears to be more rapid in Artesian wells of very
small depth ; — although, in this respect, the very deep wells of Monte Massi in
Tuscany, and Neuffen in the north-western part of the Swabian Alps, form
remarkable exceptions.

(37) p. 38. — Quetelet, in the Bulletin de 1'Acad. de Bruxelles, 1836, p. 75.

(ffl) p. 38. — Forbes, Exper. on the Temperature of the Earth at different
Depths, in the Transactions of the Royal Society of Edinburgh, vol. xvi. 1849,
Pt. II. p. 189.

("•) p. 39. — All the figures relating to the temperature of the Caves de 1'Ob-
servatoire are taken from Poisson 's The'orie Mathe'matique de la Chaleur, p. 415
and 462. On the other hand, the Annuaire mete'orologique de la France of Mar-
tins and Haeghens, 1849, p. 88, contains different corrections of Lavoisier's sub-
terranean thermometer, by Gay-Lussac. On the mean of three readings (June to
August), that thermometer gave 12°-193, when Gay-Lussac found the tempera-
ture ll°-843 ; the difference therefore being 0°'350.

(*°) p. 39. — Cassini, in the Mem. de 1'Acad. des Sciences, 1786, p. 511.

(41) p. 41. — Boussingault, " Sur la profondeur a laquelle on trouve dans la
Zone torride la Couche de Temperature invariable," in the Annales de Chimie et
de Physique, t. liii. 1833, p. 225 — 247. John Caldecott, astronomer to the
Rajah of Travancore, and Captain Newbold in India, have objected to the me-
thod recommended in this memoir, and confirmed by so many exact results
obtained in South America. Mr. Caldecott found at Trevandrum (Edinb. Trans.

XIV NOTES.

vol. xvi. Pt. III. p. 379—393), the temperature of the ground at a depth of
three feet and more (deeper, therefore, than Boussinganlt prescribes), 85° and
86° Faht. ; the mean temperature of the air being stated to be 80°'02 Faht.
Newbold's experiments (Phil. Trans, for 1845, Pt. I. p. 133) gave, at a
depth of one foot, at Bellary (in 15° 5' lat.), 4° Faht. increase of temperature
from sunrise to two hours after noon ; but at Cassargode (lat. 12° 29'), with a
clouded sky, only 1£°. Were the thermometers duly protected, and the ground
adequately screened from the direct influence of the sun's rays ? Compare
also J. D. Forbes, Exper. on the Temp, of the Earth at different Depths, in the
Edinb. Trans, vol. xvi. Pt. II. p. 189. Colonel Acosta, the meritorious his-
torian of New Granada, has recently made at Guaduas, on the south-western
declivity of the highlands of Bogota, where the mean temperature of the year is
74'84 F. at a depth of one foot, and in a covered space, a long series of observa-
tions which fully corroborate Boussingaulf's statement. Boussingault writes:
" Les observations du Colonel Acosta, dont vous connaissez la grande pre'cision
en tout ce qui inteYesse la Me'te'orologie, prouvent qne, dans Us conditions
d'dbri, la tempe'rature reste constante entre les tropiques a une tres-petite
profondeur."

(42) p. 41.— On Gualgagoc (or Minas de Chota) and Micuipampa, see Hum-
boldt, Recueil d'Observ. astron. vol. i. p. 324.

(4S) p. 41. — Essai politique sur le Roy. de la Nouv. Espagne, (2e eU) t. iii.
p. 201.

(") p. 43. — C. Von Baer, in MiddendoriFs Sibirischer Reise, Bd. i. S. 7.

(4S) p. 44. — The Merchant Fedor Schergin, head of the comptoir of the Rus-
so- American Trading Company, began, in the year 1828, to have a well sunk in
the court of a house belonging to the Company. As down to the depth of ninety-
six feet, which he reached in 1830, only frozen earth was found, and no water, he
gave up the work, until Admiral Wrangel passed through lakutsk on his way
to Sitka, and, perceiving the great scientific interest which attached to piercing
through the subterranean icy stratum, asked Mr. Schergin to continue the bor-
ing. In 1837, a depth of fully 382 Eng. feet had been reached, the ice still
continuing.

(«) p. 45.— Middendoi-ff, Reise in Sib. Bd. i. S. 125—133. " If," says
Middendorff, " we exclude all cases where the depth reached falls short of 100
feet, because, according to experience hitherto in Siberia, all lesser depths are
subject to annual variations of temperature, there still remain such anomalies or
differences in the rate of increase of temperature that, while between 150 and
200 feet, it has been found so rapid as to give only fi6 English feet for an
increase of 1° Reaum. ; between 250 and 300 feet, the observations would give

NOTES. XV

217 such feet to one such degree. We are, therefore, induced to consider that
the results of the observations hitherto obtained in the Schergin boring, or shaft,
are by no means sufficient to determine, with certainty, the measure of the in-
crease of temperature; but yet, that (notwithstanding great deviations which
may be occasioned by differences in the heat-conducting power of the strata,
and in the disturbing influence of water or air which may sink down from the
surface) we may venture to say that, on the whole, the temperature increases
1° Reaum. for a descent of not more than 100 to 117 Eng. feet." The result
of 117 Eng. feet, is the mean of six partial results, or augmentations of tem-
perature from at every 50 feet between 100 and 382 feet depth in the shaft. If
I compare the mean atmospheric temperature of the year at lakutsk (— 80<13
Reaum.) with the mean temperature given by observation of the ice at the
lowest part of the shaft (— 2°'40 Reaum.— greatest depth 382 Eng. feet),
I find 66| Eng. feet for 1° Reaum. The comparison of the deepest part with
the temperature 100 feet below the surface, gives 100 feet. From the saga-
cious numerical investigations of Middendorff and Peters on the rate of propaga-
tion of variations of atmospheric temperature, and on the maxima and minima
of temperature (Middend. S. 133 — 157 and 168 — 175), it follows, that in the
different borings, at the small depths of 7 to 20 or 21 feet in the upper portions
of the borings, there is " an increase of temperature from March to October, and
a decrease from November to April, because spring and autumn are the seasons
at which the changes of atmospheric temperature are most considerable." (S. 142
and 145.) Even carefully covered-over mines in Northern Siberia cool gradu-
ally in the course of years, from the contact of the air with the sides of the
shafts. In the Schergin shaft, however, in the course of eighteen years, this
contact has occasioned scarcely half a degree Cent. (0'9 Faht.) diminution of
temperature. A remarkable, and, as yet, unexplained phenomenon, which has
also presented itself in the Schergin shaft, is an increase of temperature which
has sometimes been perceived in winter, exclusively in the lower strata, and
"without any demonstrable external influence." (S. 156 and 178.) A more ex-
traordinary circumstance, as it appears to me, is that in the boring at Wedensk,
on the Pa'sina, with an atmospheric temperature of — 28° Reaum., — 2°'5 was
found at the very small depth of between 5 and 8i feet ! The isogeothermal
lines, to the direction of which Kupffer's sagacious investigations first called at-
tention (Kosmos, Bd. i. S. 445 ; English edition, Note 201), will probably long pre-
sent many unresolved problems. The solution is particularly difficult where
the complete piercing through the stratum of ground ice is a work of much
time. The ground ice at lakutsk can no longer be looked upon as a mere local
phenomenon, as in the view of those who attributed it to a sinking of the

XVI NOTES.

earthy strata through the agency of water, or a precipitation of earth from
water. (Middend. S. 167.)

(47) p. 45.— Middendorff, Bd. i. S. 160, 164 and 179. In these numerical
estimates and conjectures as to the thickness of the ground ice, an increase of
temperature in arithmetical progression with the depth is assumed. Whether, at
greater depths, the rate of increase of temperature may become slower, is theo-
retically uncertain ; and it is, therefore, not advisable to enter into calculations
respecting the temperature at the centre of the Earth in current-exciting molten
masses of various kinds.

(*) p. 46. — Schrenk's Eeise durch die Tundern der Samojeden, 1848, Th. i.
S. 597.

C49) p. 46. — Gustav. Rose, Reise nach dem Ural, Bd. i. S. 428.

(**) p. 47. — Compare G. von Helmersen's experiments on the relative con-
ducting power of different rocks (Me'm. de 1'Acad^mie de St. Pdtersbourg ; Me'-
langes physiques et chimiques, 1851, p. 32).

(51) p. 47.— Middendorff, Bd. i. S. 166 compared with S. 179. " The limit
of the ground ice appears to have, in Northern Asia, two southerly inflexions,
one not deep on the Obi, and one very considerable on the Lena. The line
runs from Beresow on the Obi towards Turuchausk on the lenissei, then between
Witimsk and Oleminsk, on the right bank of the Lena, and then, eastward,
ascending to the north."

(52) p. 50. — The principal passage in which the magnetic chain of rings is
spoken of is in the Platonic Ion, p. 533 D, E. ed. Steph. Subsequently, this
propagation of the attracting influence was mentioned by Pliny (xxxiv. 14) and
Lucretius (vi. 910), and also by Augustine (de Civitate Dei, xx. 4) and Philo
(de Mundi Opificio, p. 32 D. ed. 1691).

(*) p. 51.— Kosmos, Bd. i. S. 194 and 435, Anm. 32; Bd. ii. S. 293—
295, 307—322, 468, Anm. 59; and S. 481— 482, Anm. 91—93.

(54) p. 51.— Compare Humboldt, Asie Centrale, t. i. p. xl.— xlii., and Examen
crit. de 1'Hist de la Geographic, t. iii. p. 35. Eduard Biot, who, partly alone
and partly with the help of Stanislas Julien, has confirmed and enlarged, by
laborious bibliographic studies, Klaproth's researches on the antiquity of the
use of the magnetic needle in China, cites an older tradition, which, however, is
first found mentioned by writers of the first centuries of the Christian era, ac-
cording to which magnetic cars would have been used as early as under the
Emperor Hoang-ti. This celebrated monarch is supposed to have reigned 2600
years before our era (t. e. a thousand years before the expulsion of the Hyksos
from Egypt). Ed. Biot, " Sur la direction de 1'Aiguille aimantee en Chine," in
the Comptes rendus de 1'Acad. des Sciences, t. xix. 1844, p. 362.

NOTES. XVll

(55) p. 52.— Kosmos, Bd. i. S. 194 and 435, Anm. 31 (English edition,
p. 176 and Note 161). Aristotle himself (de Anima, i. 2) speaks only of
the " animation" of the loadstone as an opinion of Thales. Diogenes Laertius,
however, distinctly extends the opinion to amber, saying, " Aristotle and Hippias
maintain from Thales' doctrine," &c. The sophist Hippias of Elis, who
thought he knew everything, occupied himself with natural philosophy, and
thus with the most ancient traditions of the physiologic school. The " attracting
breath of air" which, according to the Chinese physicist Kuopho, " blows or
breathes through loadstone and amber," reminds us, according to Buschmann's
linguistic Mexican investigations, of the Aztec name for the magnet : " tlaibio-
anani tetl," signifying "the stone which draws to itself by breath" (from
" ihiotl," breath ; and " ana," draw).

(56) p. 52. — The information respecting this remarkable apparatus, extracted
. by Klaproth from Penthasaoyan, is found in greater detail in Mung-khi-pi-than;

Comptes rendus, t. xix. p. 365. Why is it that in the last-named memoir, as
also in a Chinese Herbal, it -is said that the cypress points to the west and the
magnetic needle points to the south ? Is it intended to refer to a more or less
luxuriant development of the branches according to " aspect" as respects the
sun or some prevailing wind?

(57) p. 58. — Kosmos, Bd. ii. S. 469 — 472 (English edition, Note 405). In
the time of Edward III. of England, when, as Sir Nicholas Harris has shown
(History of the Royal Navy, 1847, vol. ii. p. 180), it was customary to sail by
the compass, then called " sailstone dial," " sailing needle," or " adamant," we
see that in the expenses of fitting out "the king's ship the George," in the year
1345, mention is made of sixteen "hour-glasses" bought in Flanders. This,
however, is by no means a proof of the use of the log. Hour-glasses (the " ampol-
letas" of the Spaniards) were used, as is evident from the statements of Enciso
in Cespedes, long before the employment of the log, " Echando punto por fan-
tasia" in the " corredera de los perezosos," i. e. without casting out a log.

(5S) p. 58. — Compare Kosmos, Bd. i. S. 427, Anm. 11, and 429, Anm. 14
(English edition, Notes 141 and 144); Bd. ii. S. 373. 381, 382, and 515,
Anm. 70 — 72, and 517, Anm. 88 (English edition, p. 331, 332, and 339.
340, Notes 510. 512. and 528).

(59) p. 59. — Compare Gilbert, Physiologia nova de Magnete, lib. iii. cap. 8,
p. 124. That magnetism can be imparted to iron for a considerable time, was
said in a general way by Pliny, but without mentioning rubbing (Kosmos, Bd. i.
S. 430, Anm. 19; English edition, Note 149). Gilbert's derisive remark on
the " vulgaris opinio de montibus magneticis aut rupe aliqua magnetica, de
polo phantastico a polo mundi distante," is worthy of notice. The alterations

VOL. IV. I

XVlil NOTES.

and progressive movement which take place in the magnetic lines were still
wholly unknown to him : " varietas uniuscujusque loci constaus est."

(60) p. 59. — Historia natural de las Indias, Kb. i. cap. 17.

(61) p. 59.— Kosmos, Bd. i. S. 189; English edition, p. 171.

(82) p. 60. — I have shown, from very careful observations of the Inclination
made by myself in the Pacific, what are the conditions under which the mag-
netic Inclination may be of important practical use in determining the latitude
during the prevalence of the "garua"or fog, which, at certain seasons on the
Peruvian coast, conceals the sun and stars. (Kosmos, Bd. i. S. 185 and 428,
Anm. 14; English edition, p. 168 and Note 144.) The Jesuit Cabius, the au-
thor of the " Philosophia magnetica" (in qua nova quaedam pyxis explicatnr,
quae poll elevationem ubique demonstrat), directed attention to this subject
in the first half of the 17th century.

(w) p. 60. — Edmund Halley, in the Phil. Trans, for 1683 vol. xii. No. 148,
p. 216.

(w) p. 61. — Such lines, called by him "tractus chalyboeliticos," had also
been drawn by Pater Christoph Burrus, at Lisbon, in a map which he offered at
an exorbitant price to the King of Spain, as mentioned by Kircher in his Magnes,
ed. 2. p. 443. The first Variation Chart ever constructed has been already
mentioned in p. 56.

C55) p. 62. — Even twenty years after Halley had prepared, at St. Helena, his
Catalogue of Southern Stars (in which, unfortunately, there are none below the
6th magnitude), Hevelius, in the " Firmamentum Sobescianum," boasted of not
using a telescope, and of observing through longitudinal apertures. Halley had
been present at some of these observations (to the exactness- of which he gave
undue praise) when he visited Dantzic in 1679. Kosmos, Bd. iii. S. 60, 106
(Anm. 2 and 3), 154, 317, and 355 (Anm. 13); English edition, p. 43, xiv.
(Notes 91 and 92), 97, 221, and Ixxix. Note 370.

(66) p. 62. — Traces of variations of the magnetic Declination on different
days, and at different hours, had already been remarked in London by Hellibrand,
1634, and in Siam by Pater Tachard, 1682.

(67) p. 63. — Compare Kosmos, Bd. i. S. 432—435, Anm. 29; English edi-
tion, Note 159. The excellent construction of the Boussole d'Inclinaison first
made by Lenoir, according to Borda's plan, the possibility which it gave of free
and long vibrations of the needle, the very great diminution of the friction of the
axles, and the steadiness and good means of adjustment of the instrument, which
was provided with levels, first made it possible to obtain exact measurements
of the differences of the Earth's magnetic force in different zones.

C68N p. 65. — The numbers placed first in the table (as, for example, 1803—

NOTES. XIX

1806) indicate the date of the observations; the figures within brackets, which
are added occasionally after the titles of the memoirs, show the date of the pub-
lication of the observations, which was often much later.

(^ p. 69. — Malus (1808) and Arago (1811 , chromatic) polarisation of light.
SeeKosmos, Bd. ii. S. 370 (English edition, p. 329).

(70) p. 70. — Kosmos, Bd. i. S. 186 and 429, Anm. 17 (English edition,
p. 169 and Note 147).

(71) p. 72. — " Before the practice was adopted of determining absolute values,
the most generally used scale (and which still continues to be very frequently
referred to) was founded on the time of vibration observed by Mr. de Humboldt
about the commencement of the present century, at a station in the Andes of
South America, where the direction of the dipping-needle was horizontal — a con-
dition which was for some time erroneously supposed to be an indication of the
minimum of magnetic force at the Earth's surface. From a comparison of the
times of vibration of Mr. de Humboldt's needle in South America and in Paris,
the ratio of the magnetic force at Paris to what was supposed to be its minimum
was inferred (1'348); and from the results so obtained, combined with a similar
comparison made by myself between Paris and London in 1827 with several
magnets, the ratio of the force in London to that of Mr. de Humboldt's original
station in South America has been inferred to be 1'372 to I'OOO. This is the
origin of the number 1'372, which has been generally employed by British ob-
servers. By absolute measurements we are not only enabled to compare numeri-
cally with one another the results of experiments made in the most distant parts
of the globe, with apparatus not previously compared, but we also furnish the
means of comparing hereafter the intensity which exists at the present epoch
with that which may be found at future periods." Sabine, in the Manual for
the Use of the British Navy, 1849, p. 17.

(72) p. 74. — The desirability of concerted simultaneous magnetic observation
was first perceived by Celsius. Without speaking of the influence of the aurora
upon the declination, which had been discovered, and even measured, by his as-
sistant, Olav Hiorter, in March, 1741, Celsius wrote in the summer of the same
year to Graham, to invite him to join with him in examining whether certain ex-
traordinary perturbations, which the horary march of the declination needle
underwent from time to time at Upsala, would also be observable at the same
time in London. Simultaneity of disturbance would, he said, afford proof of the
cause of perturbation extending over considerable spaces of the Earth's surface,
and not being occasioned by accidental local influences. (Celsius, in Svenfka
Vetenskaps Academiens Handlingar for 1740, p. 44; Hiorter, in the same work
for 1747, p. 27.) When Arago had become aware that the magnetic perturba^

6 2

XX NOTES.

tions accompanying auroras extend to regions of the Earth where the auroral
light is not seen, he, in 1823, arranged with our common friend Kupffer for
simultaneous hourly observations being made at Paris and at Kasan, almost 47°
east of Paris. Similar simultaneous observations of the declination were set on
foot by me in 1828, by Arago, Reich, and myself, in Paris, Freiberg and Berlin.
See Poggend. Ann. Bd. xix. S. 337.

(ra) p. 81. — The memoir of Rudolph Wolf, referred to in the text, contains his
own daily observations of solar spots, from the 1st of January to the 30th of
June, 1852, and a comparison of Lament's periodical declination-variations with
Schwabe's results on the frequency of the solar spots in the interval from 1835
to 1850. This memoir WHS presented on the 31st of July, 1852, at a meeting
of the Society of Natural Sciences at Berne, while the more detailed memoir of
Colonel Sabine (printed in the Phil. Trans, for 1852, Pt. I. p. 116—121) had
been presented to the Royal Society of London at the beginning of March, and
read at the beginning of May in the same year, i.e. 1852. According to the
most recent examination of all the observations which have been made on the
solar spots, Wolf finds the mean period, between 1600 and 1852, to be ITU
years.

(74) p. 83. — Kosmos, Bd. iii. S. 400 and 419, Anm. 30 (English edition, p.
290 and civ. ; Note 489). Bismuth, antimony, silver, phosphorus, rock-salt, ivory,
wood, slices of apple, and leather, when placed in the vicinity of a powerful mag-
net, all show diamagnetic repulsion, and equatorial, L e. east and west, axiality.
Oxygen (pure, or mixed with other gases, or condensed in the interstices of char-
coal) is paramagnetic. In reference to crystallised bodies, see what has been
discovered, in relation to the position of certain axes, by the sagacious Pliicker.
(Poggend. Ann. Bd. 73, S. 178 ; and Phil. Trans, for 1851, Art. I. § 2836—
2842.) The repulsion by bismuth was first recognised, in 1778, by Brugmans,
and, at a later period, more thoroughly examined by Le Baillif in 1827, and by
Seebeck in 1828. Faraday himself (§ 2429^2431), Reich, and Wilhelm
Weber, who, since 1836, has been so uninterruptedly active in the advancement
of the knowledge of the Earth's magnetism, have brought forward the connection
of diamagnetic phenomena with those of induction. (Poggend. Ann. Bd. 73,
S. 241 and 253.) Weber tried to satisfy himself that diamagnetism has its
source in Ampere's molecular currents. (Wilh. Weber, Abhandlungen tiber
electro-dynamische Maasbestimmungen, 1852, S. 545 — 570.)

(ra) p. 84. — For producing this " polarity," the magnetic fluids in every par-
ticle of oxygen are separated by the Earth's "actio in distans," a certain amount,
in a determinate direction, and with a determinate force. Every particle of
oxygen thus represents a small magnet, and all these small magnets react upon

NOTES. XXI

each other, as well as upon the Earth ; and lastly, act in combination with the
Earth upon an imaginary needle, placed anywhere within or without the atmo-
sphere. We may compare the oxygen surrounding the Earth to an armature of
soft iron applied to a natural or to a steel magnet ; the magnet being supposed
spherical like the Earth, and the armature a hollow sphere like the envelope of
oxygen in the atmosphere surrounding the Earth. The degree of magnetic power
which each particle of oxygen can receive from the Earth's constant magnetism,
diminishes with increasing temperature and rarefaction of the gas. Now as the
sun's course from east to west is followed by a constant increase of temperature
and expansion, this must produce a modification in the magnetic relations of the
Earth and its oxygenous envelope, which, in Faraday's opinion, is the source of a
part of the variations of the observed elements of terrestrial magnetism. Plucker
finds that, inasmuch as the degree of force with which a magnet acts upon oxy-
gen is proportioned to the density of that gas, the- use of the magnet affords a
simple eudiometric means of recognising the presence of one or two hundredth
parts of free oxygen in a mixture of gases.

(76) p> 86.— Kosmos, Bd. iv. S. 10 and 11 (English edition, p. 10 and 11).

(") p. 87.— Kepler, in Stella Martis, p. 32 and 34. Compare therewith his
" Mysterium cosmogr." cap. 20, p. 71.

(78) p. 87.— Kosmos, Bd. iii. S. 416, Anm. 23 (English edition, Note 482),
where, both in the original and in the translation, " Basis Astronomiae " has been
printed instead of, as it ought to be, " Clavis Astronomic." The passage
(§ 226) in which the solar evolution of light is termed " a perpetual aurora," is
not to be found in the first edition of the Clavis Astr. of Horrebow (Havn. 1730),
but only in the later edition, contained in Horrebow's " Operum mathematico-
physicorum," t. i. Havn. 1740, pag. 317, where there is a Second Part of the
Clavis in which the passage occurs. Compare with Horrebow's view the entirely
accordant views of Sir William and Sir John Herschel, Kosmos, Bd. iii. S. 45, 56
(Anm. 22), 256 and 262 (English edition, p. 35, xi., Note 68), 176, and Note
289.

(ra) p. 87.— Me'moires de Mathe'm. et de Phys. prdsente's a 1'Acad. Eoy. des
Sc. t. ix. 1780, p. 262.

(M) p. 88. — " So far as these four stations (Toronto, Hobarton, St. Helena, and
the Cape), so widely separated from each other and so diversely situated, justify
a generalisation, we may arrive at the conclusion, that at the local hour of seven
and eight A.M. the magnetic declination is everywhere subject to a variation of
which the period is a year, and which is everywhere similar in character and
amount, consisting of a movement of the north end of the magnet from east to
west between the northern and the southern solstice, and a return from west to

XX11 NOTES.

east between the southern and the northern solstice, the amplitude being about
five minutes of arc. The turning periods of the year are not, as many might
be disposed to anticipate, those months in which the temperature at the surface
of our planet, or of the subsoil, or of the atmosphere (as far as we possess the
means of judging of the temperature of the atmosphere) attains its maximum
and minimum. Stations so diversely situated would indeed present in these
respects thermic conditions of great variety: whereas uniformity in the epoch of
the turning periods is a not less conspicuous feature in the annual variation than
is similarity of character and numerical value. At all the stations the solstices
are the turning periods of the annual variation at the hour of which we are
treating. The only periods of the year in which the diurnal or horary variation
at that hour does actually disappear, are at the equinoxes, when the sun is
passing from the one hemisphere to the other, and when the magnetic direction
in the course of its annual variation from east to west, or vice versa, coincides
with the direction which is the mean declination of all the months and of all the
hours. The annual variation is obviously connected with, and dependent on,
the Earth's position in its orbit relatively to the sun, around which it revolves; as
the diurnal variation is connected with, and dependent on, the rotation of the
Earth on its axis, by which each meridian successively passes through every
angle of inclination to the sun in the round of twenty-four hours." Sabine, On
the annual and diurnal Variations (Observations made at the Magn. and Meteonol.
Observatory at Toronto, vol. ii. p. xvii — xx.). Compare also his Memoir on the
annual Variation of the magnetic Declination at different Periods of the Day, in
the Phil. Trans, for 1851, Pt. II. p. 635, and the Introduction to vol. i. of the
Observations made at the Observatory at Hobarton, p. xxxiv — xxxvi.

(81) p. 88. — Sabine, On the Means adopted for determining the absolute Values,
secular Change, and annual Variation of the terrestrial magnetic Force, in the
Phil. Trans, for 1850, Pt. I. p. 216. Also in Sabine's Opening Address to the
Meeting of the British Association at Belfast, in 1852, he says: — " It is a re-
markable fact, which has been established, that the magnetic force is greater in
both the northern and southern hemispheres in the months of December, January,
and February, when the sun is nearest to the Earth, than in those of May, June,
and July, when he is most distant from it : whereas, if the effects were due to
temperature, the two hemispheres should be oppositely instead of similarly
affected in each of the two periods referred to.

(82) p. 88. — Lamont, in Poggend. Ann. Bd. 84, S. 579.

(") p. 89. — Sabine, On periodical Laws discoverable in the mean Effects of the
larger magnetic Disturbances, in the Phil. Trans, for 1852, Pt.I. p. 12 1. (Kos-
mos, Bd. iv. S. 73; English edition, p. 78.)

NOTES. XX111

(**) p. 89.— Kosmos, Bd. iii. S. 402 (English edition, p. 292).
C85) p. 89.— Kosmos, Bd. iii. S. 238 (English edition, p. 158).
(M) p. 90. — Kreil, Einfluss des Mondes auf die magnetische Declination,
1852, S. 27, 29 and 46.

(87) p. 91.— Kosmos, Bd. i. S. 407, Anm. 55, and, in regard to meteoric
stones, S. 137; also Bd. iii. S. 594 (English edition, Vol. i. Note 85, and p.
122; Vol. iii. p. 421).

(88) p. 92. — Compare Mrs. Somerville, in her brief but lucid representation of
terrestrial magnetism, founded on Sabine's writings, in vol. ii. p. 102 of her
Physical Geography (2nd edition). Sir James Koss, who, on his great antarctic
expedition, crossed this curve of weakest force, December, 1839, lat. 19° S., and
long. 29° 13' W., and who has the great merit of having been the first to deter-
mine its place in the southern hemisphere, terms it the " equator of least in-
tensity." See his Voyage to the Southern and' Antarctic Regions, vol. i.
p. 22.

(") p. 92. — " Stations of an intermediate character situated between the
northern and southern magnetic hemispheres, partaking, although in opposite
seasons, of those contrary features which separately prevail (in the two hemi-
spheres) throughout the year." Sabine, in the Phil. Trans, for 1847, Pt. I.
p. 53 and 57.

(") p. 93. — The Pole of Intensity of Force is not the Pole of Verticity.
(Phil. Trans, for 1846, Pt. III. p. 255.)

(9I) p. 93. — Gauss, allgem. Theorie des Erdmagnetismus, § 31.

(ffi) p. 94. — Phil. Trans, vol. xxxiii. for 1724, 1725, p. 332 ("to try if
the dip and vibrations were constant and regular ").

(93) p. 94. — Novi Comment. Acad. Scient. Petropol. t. xiv. pro anno 1769,
Pars II. p. 33. See also Le Monnier, Lois du Magrietisme compare'es aux Ob-
servations, 1776, p. 50.

(M) p. 95. — The sign + prefixed to a latitude signifies " North," and the
sign - " South" [In this translation the words North and South, or the letters
N. and S., are employed for the same purpose. — Ed.] ; and East and West signify
East and West longitude from Paris, unless otherwise stated. [In this translation,
on the other hand, East and West, or E. and W., signify East or West longitude
from Greenwich, unless otherwise stated.— Ed.] In the section on Terrestrial
Magnetism, from p. 80 to p. 153, wherever marks of quotation (" ") occur with-
out any particular memoir of Colonel Sabine's being expressly referred to, either
in the text or in the notes, it is to be understood that the passage is taken from
my friend's manuscript communications to myself.

(95) p. 96. — Fifth Eeport of the British Association, p. 72; Seventh Report,

XXIV NOTES.

p. 64 and 68; Contributions to Terrestrial Magnetism, No VII. in the Phil.
Trans, for 1846, Pt. III. p. 254.

(*) p. 97.— Sabine, in the Seventh Report of the Brit. Assoc. p. 77.

(97) p. 97. — Sir James Ross, Voyage in the Southern and Antarctic Regions,
vol. i. p. 322. This great navigator crossed the line of greatest force twice
between Kerguelen and Van Diemen's Islands; first in 46° 44' S. lat., 128° 28'
E. long., where the force was 2'034, from whence it diminished to T824 at
Hobarton (vol. i. p. 103 and 104); and a year afterwards, from the 1st of
January to the 3rd of April, 1841, when we find, from the journals of the
Erebus, that from 77° 47' to 51° 16' South latitude, and 175° 43' to 136° 52'
East longitude from Greenwich, the intensity of the Earth's magnetic total
force was always above 2'00, and even reached 2'07. (Phil. Trans, for 1843,
Pt. II. p. 211 — 215.) Sabine's result for the one focus of the southern hemi-
sphere (lat. 64°, long. 137° 30' E.), which I have given in the text, is derived
from Sir James Ross's observations from the 19th to the 27th of March, 1841,
between 58° and 64° 26' S. lat., and 128° 40' and 148° 20' E. long., " crossing
the southern isodynamic ellipse of 2'00 about midway between the extremities
of its principal axis." Contributions to Ter. Mag. in the Phil. Trans, for 1846,
Pt. III. p. 252.

(") p. 97. — Ross's Voyage, vol. ii. p. 224. According to the instructions
given for the voyage, the two southern foci of maximum force were conjectured
to be in 47° S. lat. 140° E. long., and in 60° S. lat. 235° E. long.

(99) p 98.— Phil. Trans, for 1850, Pt. I. p. 201 ; Admiralty Manual, 1849,
p. 16; Erman, Magnet. Beob. S. 437—454.

(10°) p. 99. — On the map of the isodynamic lines in North America, in No.
VII. of Sabine's Contributions to Terrestrial Magnetism, by an accidental error,
14'88 is the number given instead of 14'21; but the latter, which is the true
number, is given in the text of the Memoir, p. 252. There is also 'a typo-
graphical error in the portion added to Note 158 in the English translation of
Kosmos (Vol. i. p. 414), 13'9 being printed instead of, as it ought to be, 14'21.

(l01) p. 99. — 15-60 is taken from No. VII. of Sabine's Contrib. p. 252.s
From the Table of the Observations of the Erebus (Phil. Trans, for 1843,
Pt. II. p. 169 and 172), we see that some of the observations, taken on the ice
in 77° 47' S. lat., 172° 40' W. long., even gave 2'124 in the arbitrary scale.
The value of 15'60 in absolute scale supposes the intensity of the force at
Hobarton to be 13*51 ; the latter value being to be regarded as provisional
(Mag. and Met. Observations at Hobarton, vol. i. p. Ixxv.), and having indeed
been a little augmented in vol. ii. p. xlvi., where it is given at 13' 5 6. In the

NOTES. XXV

Admiralty Manual, p. 17, I find the force at the stronger southern focus altered
accordingly to 15'8.

(102) p. 99. — Sabine, in the English translation of Kosmos, Vol. i. p. 414.

(103) p. 100. — See the interesting representation entitled: Map of the World,
divided into hemispheres by a plane coinciding with the meridians of 100 and
280 E. of Greenwich, exhibiting the unequal distribution of the magnetic in-
tensity in the two hemispheres, Plate V., British Association Reports for 1837,
p. 72 — 74. Erman found the force almost always below 0'76 (very weak
therefore) between 24° 25' and 13° 18' S. lat., and between 34° 52' and 32° 42'
W. long, from Greenwich.

(104) p. 100. — Kosmos. Bd. i. S. 193 and 435. Anm. 30 (English edition,
p. 175 and Note 160).

(105) p. 100.— Ross's Voyage in the Southern Seas, vol. i. p. 22 and 27. See
also Kosmos, Bd. iv. S. 84, Anm. 88 (English Edition p. 92, and Note 88).

(106) p. 100. — See Table of Sulivan's and Dunlop's Observations in Phil.
Trans. 1840, Pt. I. p. 143. Their minimum, however, was not lower than
8-00.

(107) p. 101. — We obtain 1:2'44 by comparing, in the absolute scale, 6'4 at
St. Helena with 15- 60 at the stronger southern focus; or 1 : 2'47 by comparison
of St. Helena with the southern maximum augmented to 15'8 (Admiralty
Manual, p. 17); 1 : 2'91 by comparing, in the relative scale, Erman's observa-
tion in the Atlantic (0706) with the southern focus (2'06); and even 1 : 2-95,
if we compare, in the absolute scale, the weakest force observed by Erman (5*35)
with the highest result for the southern focus (15'8). A mean number would
be 1 : 2'69. Compare for the intensity of the force at St. Helena (6'4 in abso-
lute, or 0'845 in relative scale) the earliest observations by Fitz Roy (0'836),
Phil. Trans, for 1847, Pt. I. p. 52, and British Association Report for 1837,
p. 56.

(108) p. 101. — Compare English translation of Kosmos, Vol. i. Editor's Note
158 p. liii. — Iviii., and No. VII. of Contributions to Terrest. Magnetism, p. 256.

(loe) p. 102. — What can have led to the erroneous result obtained in the
coal mines of Flenu, viz., that at a depth of 83 French feet the horizontal in-
tensity increased O'OOl? Journal de 1'Institut, 1845, Avril, p. 146. In an
English coal mine, at a depth of a thousand feet, Kenwood found no increase of
force. (Brewster, Treatise on Magnetism, p. 275.)

(»°) p. 103.-Kosmos, Bd. i. S. 418, Bd. iv. S. 36 (English edition, Vol. i.
Note 124; Vol. iv. p. 36).

(m) p. 103. — In my observations, a diminution of magnetic force with in-

XXVI NOTES.

creasing height results from the comparison of the Silla de Caracas (8105 feet
above the sea; force 1-188) with the harbour of La Guayra (height 0; force
1-262), and with the town of Caracas (height 2484 feet; force 1-209); from
the comparison of the town of Santa F<$ de Bogota (height 8190 feet; force
1-147) with the Chapel of Nuestra Senora de Guadalupe (height 10,128 feet.
force 1-127), placed, like a swallow's nest, on a steep precipice overhanging the
town, and in the closest proximity to it; from the comparison of the volcano of
Purace (height 13650 feet; force 1-077) with the mountain-village of Pnraee
(height 8136 feet; force 1-087), and with the neighbouring town of Popayan
(height 5466 feet; force 1'117); and from the comparison of the town of Quito
(height 8952 feet; force 1-067) with the village of San Antonio de Lulum-
bamba (height 7650 feet; force 1*087), situated in an adjacent ravine, and
exactly on the geographical equator. An opposite result was derivable from the
most elevated site at which I ever made magnetic experiments of vibration, i. e.
at a height of 14,960 feet on the side of the long-extinct volcano of Antisana,
opposite Chussulongo. The observations had to be made in a large cave, and the
great increase of force was no doubt caused by a local attraction of the rock
(trachyte), as was the case in experiments made by Gay-Lussac and myself
on the margin of the crater of Vesuvius, and in the crater itself. In the cave
of Antisana I found the magnetic force raised to 1-188, it being scarcely 1-068
on the neighbouring elevated, but yet much lower, table-lands. The intensity of
the force at the Hospice on the St. Gothard (1-313) was greater than that at
Airolo (1-309), though less than at Altorf (1-322); but the force at the Ursern
Loch (1-307) was less than at Airolo. So also Gay-Lussac and myself found
the intensity of the force at the Hospice on Mont Ceiiis 1-344, that at Lans le
iourg, at the foot of Mont Cenis, 1-323, and at Turin 1-336. The greatest
anomalies, as might naturally have been anticipated, were presented by the still
active volcano of Vesuvius. When, in 1805, we found the magnetic force 1-274
at Naples, and 1-288 at Portici, we found it increased to 1-302 at the hermitage
of San Salvador, and diminished to 1-193 (lower than any place in the whole
district) in the crater of Vesuvius. Ferrugineous matter in the lavas, proximity
of magnetic poles in particular pieces, and the probably, on the whole, weakening
effect of the heating of the ground, all contribute to produce the most opposite
local disturbances. Compare my Voyage aux Kegions dquinoxiales, t. iii. p.
619—626, and Mem. de la Socie'te' d'Arcueil, t. i. (1807) p. 17—19.

(I12) p. 103. — Kupffer's observations were not made at the summit of Mount
Elbrouz, and only apply to the difference of elevation (4500 feet) between two
stations, the Bridge of Malpa, and the slope of the Mountain of Kharbis, situated
at no inconsiderable distance from each other in a horizontal line. On the ob-

NOTES. XXV11

jections which Necker and Forbes have raised as to the result, see vol. xiv. of
the Transactions of the Eoyal Society of Edinburgh, 1840, p. 23 — 25. [All the
elevations given in this and the preceding note are, as in the original, in French
feet; it did not appear necessary to encumber the page by stating them in Eng-
lish feet likewise.— Ed.]

(113) p. 104. — Compare Laugier and Mauvais, in the Comptes rendus,t. xvi.
1843, p. 1175, and Bravais, Observ. de I'lntensite* du Magne'tisme terrestre en
France, en Suisse et en Savoie, in the Annales de Chimie et de Phys. 39 se'rie,
t. xviii. 1846, p. 214; Kreil, Einfluss der Alpen auf die Intensitat, in the
Denkschriften der Wiener Akad. der Wiss., mathem. naturwiss. Cl. Bd. i. 1850,
S. 265, 279, and 290. It is the more remarkable that a very exact observer,
Que'telet, should have found, in 1830, the horizontal intensity at Geneva 1*080,
at the Col de Balme 1-091, and at the Hospice on the Grand St. Bernard 1*096,
being an increase with increasing elevation. Comp. Sir David Brewster, Treatise
on Magnetism, p. 275.

(1H) p. 104.— Annales de Chimie, t. lii. (1805) pp. 86 and 87.

(115) p. 104. — Arago, in the Annuaire du Bureau des Longitudes pour 1836,
p. 287. Forbes, in the Edinb. Trans, vol. xiv. (1840) p. 22.

(116) p. 105. — Faraday, Exper. Researches in Electricity, 1851, p. 53 and
77, § 2881 and 2961.

(m) p. 106.— Christie, in the Phil. Trans, for 1825, p. 49.

(118) p. 106. — Sabine, On the periodical Laws of the larger magnetic Disturb-
ances, in the Phil. Trans, for 1851, Pt. I. p. 126; and the same, On the annual
Variation of the magn. Declin., in the Phil. Trans, for 1851, Pt. II. p. 636.

(119) p. 107. — Observ. made at the Magn. and Meteor. Observatory at Toronto,
vol. i. (1840—1842) p. Ixii.

(12°) p. 107. — Sabine, in Magn. and Meteor. Observations at Hobarton, vol.i.
p. Ixix. " There is also a correspondence in the range and turning hours of the
diurnal variation of the total force at Hobarton and at Toronto, although the
progression is a double one at Toronto, and a single one at Hobarton." The time
of the maximum force is, at Hobarton, between 5 and 6 P.M., and that of the
principal maximum at Toronto 6 P.M. ; the time of the minimum at Hobarton is
between 8 and 9 A.M., and that of the secondary minimum at Toronto at 10 A.M.
Therefore, according to local time, the increase and decrease of the force follow
the same hours in the two hemispheres; not the opposite ones, as in the case of
the inclination and declination. On the causes of this phenomenon see p. Ixix.
of the above-cited work. (Compare also Faraday, On Atmospheric Magnetism,
§ 3027—3034.)

(m) p. 107. — Phil. Trans, for 1850, Pt. I. p. 215 to 217; Magnetical Ob-

XXV111 NOTES.

servations at Hobarton, vol. ii. (1852) p. xlvi. Compare what has been said
above in Note 81. " The total force shows less difference at the opposite seasons
of the year at the Cape of Good Hope than does the inclination." Magn. Observ.
Cape of Good Hope, vol. i. (1851) p. Iv.

(122) p. 108. — See the magnetic portion of my Asie Centrale, t. iii. p. 442.

(l83) p. 108. — Sir John Barrow, Arctic Voyages of Discovery, p. 521 and
529.

(1M) p. 108. — The highest inclination yet observed in Siberia was only 82°
16'; by Middendorff, on the River Taimyr, in 74° 17' N. lat, and 95° 42' E.
long. (Middend. Sibir. Reise, Th. i. S. 194).

(12S) p. 108. — Sir James Ross, Voyage to the Antarctic Regions, vol. i.
p. 246. " I had long cherished the ambitious hope to plant the flag of my
country on loth the magnetic poles of our globe; but the obstacles which pre-
sented themselves being of so insurmountable a character, was some degree of
consolation, as it left us no grounds for self-reproach." (p. 247.)

(186) p. 109. — Sabine, Pendulum Experiments, 1825, p. 476.

(m) p. 109. — Sabine, in the Phil. Trans, for 1840, Pt. I. p. 137, 139 and
146. For the movement of the African node, I follow the map appended to
that memoir.

(128) p. 110. — I subjoin, as is always my custom, the elements of this not
unimportant determination: Micuipampa, a small mountain-town in Peru, at
the foot of the Cerro de Gualgayoc; 6° 44' 25" S. lat., 80° 53' 3" W. long, from
Paris; elevation 11, 140 French feet above the Pacific; magnetic inclination
0°*42 North (centesimal division of the circle). — Caxamarca, a town situated
on a plain, 8784 French feet above the sea, in 7° 8' 38" S. lat, 5h 23m 42"
W. long, from Paris; inclination 0°'15 South. — Montan, a hacienda, surrounded
by herds of lamas, in the heart of the mountains; 6° 33' 9" S. lat., 5h 26m 51s
W. long.; elevation 8042 French feet; inclination 00>70 North. — Tomependa, in
the province Jaen de Bracamoros, at the mouth of the Chinchipe at its junction
with the Amazons; in 5° 31' 28" S. lat, 80° 57' 30" W. long.; elevation 1242
French feet; inclination 30-55 North. — Truxillo, a town in Peru, on the sea-
coast, in 8° 5' 40" S. lat, 81° 23' 37" W. long.; inclination 2°'15 South.
Humboldt, Recueil d'Observ. astron. (Nivellement barome'trique et ge'ode'sique,)
vol. i. p. 316, No. 242, 244 — 254. For the bases of the astronomical deter-
minations by star-altitudes and chronometers, see the same work, vol. ii. p. 379
—391. By a singular accident, the result of my inclination observations in
1802 (7° 2' S. lat, 81° 8' W. long.), notwithstanding the effect of secular
change, agrees not badly with Le Monnier's theoretical estimation. He said,
" To the north of Lima, the magnetic equator must be found, in 1776, in 7i°, or

NOTES. XXIX

at most in 6^° South latitude!" (Lois du Magnetisms comparers aux Observa-
tions, Partie II. p. 59.) [In this Note the longitudes have not, as elsewhere,
been converted into longitudes from Greenwich ; they are all from Paris. — Tr.]

(129) p. 110. — Saigey, Mem. sur 1'Equateur magne'tique d'apres les Observ.
du Capitaine Duperrey, in the Annales Maritimes et Coloniales, Dec. 1833, t. iv.
p. 5. It was already remarked, in this memoir, that the magnetic equator is
not an isodynamic line, but that the intensity of the magnetic force varies, at
different parts of the line of no dip, from 1 to 0'867.

(13°) p. 110. — This position of the magnetic equator has been confirmed, for
1830, by Erman. On his return from Kamtschatka to Europe, he found the
inclination almost null in 1° 30' S. lat, 134° 57' W. long.; in 1° 52' S. lat.,
137° 30' W. long.; in 1° 54' N. lat., 136° 5' W. long.; and in 2° 1' S. lat,,
141° 28' W. long. (Erman, Magnet. Beob. 1841, S. 536.) [In this Note, as
in Note 128, the longitudes have all been left in longitudes from Paris, instead
of being converted into longitudes from Greenwich, as is done wherever nothing
is stated to the contrary. — Ed.]

(131) p. 110. — Wilkes' United States Exploring Expedition, vol. iv. p. 263.

(132) p. 111.— Elliot, in the Phil. Trans, for 1851, Pt. I. p. 287—331.
(IM) p. 111. — Duperrey, in the Comptes rendus, t, xxii. 1846, p. 804—806.
(m) p. 113.— Letter from Arago to myself from Metz, 13th December 1827:

" J'ai parfaitement constate*, pendant les aurores boreales qui se sont montrees
dernierement a Paris, que 1'apparition de ce phe'nomene est toujours accompagnee
d'une variation dans la position des aiguilles horizontals et d'indinuison comme
dans 1'intensite. Les changements d'inclinaison ont ete' de 7' k 8'. Par cela
seul 1'aiguille horizontale, abstraction faite de tout changement d'intensite', devait
osciller plus.ou moins vite suivant 1'e'poque ou se faisait 1'observation; mais en
corrigeant les resultats par le calcul des effets imme'diats de 1'incliijaison, il m'est
encore reste une variation sensible d'intensite. En reprenant, par une nouvelle
methode, les observations diurnes d'inclinaison dont tu m'avais vu occupe pen-
dant ton dernier sejour a Paris, j'ai trouve', non par des moyennes, mais chaque
jour, une variation reguliere : 1'inclinaison est plus grande le matin a 9h que le
soir a 6h. Tu sais que 1'iutensite', mesuree avec une aiguille horizontale, est au
contraire a son minimum k la premiere epoque, et qu'elle atteint son maximum
entre 6h et 7h du soir. La variation totale e'tant fort petite, on pouvait supposer
qu'elle n'e'tait dfte qu'au seul changement d'inclinaison; et en effet la plus grande
portion de la variation apparente d'intensite depend de 1'alte'ration diurne de la
composante horizontale: mais, toute correction faite, il reste cependant une
petite quantite comme indice d'une variation reelle d'intensite" From another
letter from Arago, written from Paris, March 1829, a short time before my

XXX NOTES.

Siberian journey: " Je ne suis pas e'tonne' que tu reconnais avec peine la varia-
tion diurne d'inclinaison dont je t'ai parte, dans les mois d'hiver; c'est dans les
mois chauds seulement que cette variation est assez sensible pour etre observed
avec une loupe. Je persiste toujours a soutenir que les changements d'inclinaison
ne suffisent pas pour expliquer le changement d'intensite' deMuit de 1'observation
d'une aiguille horizontale. Une augmentation de temperature, toutes les autres
circonstances restant les memes, ralentit les oscillations des aiguilles. Le soir,
la temperature de mon aiguille horizontale est toujours superieure k la tempe'ra-
ture du matin; done 1'aiguille devrait,par cette cause, faire le soir, en un temps
donne', moins d'oscillations que le matin; or elle en fait plus que le changement
d'inclinaison ne le comporte: done du matin au soir il y a une augmentation
reelle d'intensite dans le magne'tisme terrestre." Later, and far more numerous,
observations at Greenwich, Berlin, St. Petersburg, Toronto in Canada, and Ho-
barton in Van Diemen's Land, have confirmed Arago's statement, made in 1827,
that the horizontal force is greater in the evening than in the morning. At
Greenwich, the principal maximum of the horizontal force is at 6h, and the
principal minimum at 22h or Oh ; at Schulzendorf, near Berlin, maximum at 8h,
minimum at 21h; at Petersburg, maximum 8h, minimum 23h 20m; at Toronto,
maximum 4h, minimum 23h; speaking always of local time at each place.
(Airy, Magn. Observ. at Greenwich for 1845, p. 13; for 1846, p. 102; for
1847, p. 241; Eiess and Moser, in Poggend. Ann. Bd. xix. 1830, S. 175;
Kupffer, Compte-rendu annuel de 1'Obs. Central Magn. de St. Pe'tersb., 1852,
p. 28; and Sabine, Magn. Obs. at Toronto, vol. i. 1840—1842, p. xlii.) At
the Cape of Good Hope, and at St. Helena, the turning hours for the horizontal
force are very different, and indeed almost opposite; the horizontal force at both
those stations being weakest in the evening. (Sabine, Magn. Obs. at the Cape of
Good Hope, vol. i. p. xl. ; at St. Helena, p. 40.) It is not, however, so through-
out the southern hemisphere. " The principal feature in the diurnal change of
the Jiorizontal force at Hobarton is the decrease of force in the forenoon, and its
subsequent increase in the afternoon." (Sabine, Magn. Obs. at Hobarton, vol. i.
p. liv.; vol. ii. p. xliii.

(135) p. 114.— Sabine, Hobarton, vol. i. p. Ixvii. and Ixix.

(136) p. 117.— Intensity of the total force at Hobarton, max. 5£h, min. 2QJh;
at Toronto, principal max. 6h, principal min. 14h; secondary max. 20h, secondary
min. 22h. Compare Sabine, Toronto, vol. i. p. Ixi. and Ixii. with Hubarton,
vol. i. p. Ixviii.

(18?) p. 1 18. — Sabine, Report on the Isoclinal and Isodynamic Lines hi the
British Islands, 1839, p. 61—63.

(138) p. 1 19. — Homboldt, in Poggend. Annalen, Bd. xv. S. 319— 33G.

s NOTES. XXXI

Bd. xix. S. 357 — 391; and in the Voyage aux Regions e'quinox. t. iii. p. 616
and 625.

(139) p. 119.— Hansteen, iiber jahrliche Veranderung der Inclination, in Pog-
gend. Ann. Bd. xxi. S. 403—429. Compare also, on the influence of the
movement of the nodes of the magnetic equator, Sir David Brewster, Treatise on
Magnetism, p. 247. Now that, through the abundance of observations at fixed
stations, we have so vast a field for special inquiries, we find fresh and fresh
complications arise in our search after regularity and law.
• (ua) p. 119.— Phil. Trans, for 1841, Pt. I. p. 35.

(H1) p. 119. — Compare Sawelieff, in the Bulletin physico-mathematiqne de
1'Acad. Imp. de St. Pdtersb. t. x. No. 219, with Humboldt, Asie Centr. t. iii.
p. 440.

(142) p. 120. — Sabine, Magn. Observ. at the Cape of Good Hope, vol. i.
p. Ixv. If we might trust La Caille's dip observations in 1751 (he, indeed, never
failed to invert the poles, but his needle did not move sufficiently freely), there
would result for the Cape an increase of inclination of 10°*08' in eighty-nine
years.

(143) p. 120. — Arago, in the Annuaire du Bureau des Long, pour 1825,
p. 285—288.

(144) p. 121. — I repeat here that all the European dips cited in this part
of the text are in the usual graduation of the circle in 360°; it is only those
observed by me on the American continent, prior to the month of June 1804
(Voy. aux Regions equinox, t. iii. p. 615 — 623), which refer to a centesimal
graduation.

(145) p. 123. — The Churprinz Mine at Freiberg in the Saxon Erzgebirge:
the subterranean place of observation was on the 7th " Gezeugstreck," in the
"Ludwiger Spathgang;" 80 "Lachter" east of the " Treibschacht," 40
"Lachter" west of the "Kunstscbacht;" depth 133^ "Lachter." The obser-
vations were made with Freiesleben and Reich, at 2^h P.M. (temperature of the
mine 15°-6 Cent.); inclination with needle A 67° 37''4; needle B 67° 32''7;
mean of the two i.eedles in the mine 67° 35''05. In the open air, at the surface
of the ground, at a spot precisely vertical above the subterranean one, at 11 AJSI.,
needle A gave 67° 33''87, and needle B 67° 32'-12; mean of the two needles
at the upper station 67° 32''99 (temperature of the air 15°'8 Cent.). Difference
between the results at the upper and lower stations 2'-06. Needle A, which,
being the stronger, was always regarded by me with most confidence, gave the
difference even higher than the mean of the two, being 3/-53 in favour of the
lower station ; the difference shown by needle B, taken alone, being scarcely
sensible. (Humboldt, in Poggend. Ann. Bd. xv. S. 326.) The method which I

XXX11 NOTES.

always employed has been described by me in detail in my Asie Centrale, p.
465 — 467. A mean result with each needle consists of sixteen readings. In
estimating the probable value of determinations of such small quantities, it is
important to consider the details of observation.

(146) p. 123.— Kosmos, Bd. i. S. 417 (English edition, Note 125).

(147) p. 123. — Humboldt, Voyage aux Regions e'quinox. t. i. p. 515 — 517.

(148) p. 125.— Erman, Reise urn die Erde,Bd. iL S. 180.

(149) p. 125. — Kosmos, Bd. iv. S. 53 (English edition, p. 55). Petrus Pere-
grini wrote to a friend that, in 1269, he had found the variation in Italy 5°
east.

(15°) p. 126. — Humboldt, Examen crit. de 1'Hist, de la Ge'ogr. t. iii. p. 29, 36,
38 and 44 — 51. Although Herrera (Dec.i. p. 23) says that Columbus had re-
marked that the magnetic variation (declination) is not the same in the day and
at night, yet this statement by no means justifies us in attributing to the great
discoverer a knowledge of the diurnal variation of the declination. From the
genuine journal of Columbus, in his Voyages published by Navarrete, we learn
from the part belonging to the 17th and 30th of September 1492, that he re-
ferred such changes to an " unequal movement of the pole-star and the pointers,"
called by him watchers or guards (guardas). (Examen crit. de 1'Hist. de la
Ge'ogr. t. iii. p. 56—59.)

(151) p. 126.— Kosmos, Bd. iv. S. 60, Anm. 66 ; and S. 70, Anm. 72
(English edition, Note 66; and Note 72). The oldest printed observations in
London are those of Graham, in the Phil. Trans, for 1724, 1725, vo). xxxiii.
p. 96 — 107. (An Account of Observations made of the Horizontal Needle,
at London, 1722 — 1723, by Mr. George Graham.) He said that the change
in the declination depended " neither upon heat, nor cold, dry, or moist air." —
" The variation is greatest between 12 and 4 in the afternoon, and least at 6
or 7 in the evening." These are not, indeed, the true turning hours.

(152) p. 127. — This is shown by numerous observations ; by those of George
Fuss and Kowanko, at the observatory belonging to the Greek convent in Pekin ;
Anikin at Nertchinsk, and Riddell at Toronto in Canada (all places having
west declination) ; also by observations made by Kupffer and Simonoff, at Ka-
san ; by Wrangel at Sitka, on the north-west coast of America, notwithstanding
the frequent auroral disturbances ; by Gilliss at Washington ; Boussingault at
Marmato, in South America ; and by Duperrey at Payta, on the coast of Peru
(all places having east declination). The mean declination was : at Pekin
(Dec. 1831), 2° 15' 42" W. (Poggend. Ann. Bd. xxxiv. S. 54); at Ner-
tchinsk (Sept. 1832), 4° 7' 44" W. (Poggend. Ann. Bd. xxxiv.- S. 61) ; at To-
ronto (Nov. 1847), 1° 33' W. (Compare Observ. at the Magn. and Met. Observa-

NOTES. XXXlll

tory at Toronto, vol. i. p. xi.; and Sabine, in the Phil. Trans, for 1851, Ft. II.
p. 636) ; Kasan (Aug. 1828), 2° 21' E. (Kupffer, Simonoff, and Erman, Reise
urn die Erde, Bd. ii. S. 532) ; Sitka (Nov. 1829), 28° 16' E. (Erman, Reise
urn die Erde, Bd. ii. S. 546) ; Marmato (Aug. 1828), 6° 33' E. (Humboldt, in
Poggend. Ann. Bd. xv. S. 331) ; Payta (Aug. 1823), 8° 56' E. (Duperrey,
in the Connaissance des Temps pour 1828, p. 252). At Tiflis the needle moves
westward from 19h to 2h (Parrot, Reise zum Ararat, 1834, Th. II. S. 58).

(153) p. 128. — Extracts from a letter written by me to Karsten, from Rome,
June 22, 1805, printed in Hansteen's Magnetismus der Erde, 1819, S. 459, on
the subject of " four movements of the magnetic needle, as it were four magnetic
ebbs and flows analogous to the barometric periods." On the subject of the so
long neglected nocturnal variations of the declination, compare Faraday, " On
the Night Episode," § 3012—3024.

(1M) p. 128. — Airy, Magn. and Met. Observations made at Greenwich, 1845
(Results), p. 6 ; 1846, p. 94 ; 1847, p. 236. I quote from a letter, dated the
llth October 1836, from my friend, the distinguished director of the Berlin
Observatory, Encke, to show how well the earliest assigned turning hours, both
of the day and of the night, derived from corresponding observations at Berlin
and Breslau, accord with those which have since been obtained from the fuller
system of observation at the Greenwich and Toronto observatories: — " I do not
think that there can be any doubt, in general, as to the existence of the nocturnal
maximum, or inflection of the curve of the diurnal variation of the declination ,
Dove also inferred it from the Freiberg observations in 1830 (Poggend. Ann.
Bd. xix. S. 373). In judging of the phenomena, graphical representations are
much more advantageous than tables of numbers. In the former, great irregu-
larities are easily seen to be such, and admit of a mean line being drawn ; while
in the latter the eye is frequently deceived, and takes a single striking irregu-
larity for a real maximum or minimum. The period appears to be determined
by the four turning hours : —

" Greatest east declination ... 20h - - principal max. east.

„ west „ - - - 1 - - principal min. east.

Secondary east „ . _ _ 10 - - minor max. east.

,, west „ - - - ] 6 - - minor min. east.

" The secondary minimum, or nocturnal elongation towards the west, falls, pro-
perly speaking, between 15h and 17h, being sometimes nearer to the one, and
sometimes to the other of those hours." It is scarcely necessary to remark,
that what Encke and I called minima to the east (at lh and 16h) are the
same which, at the British and American observatories founded in 1840, are
VOL. IV. c

XXXIV

NOTES.

called maxima to the west ; and that our " maxima to the east " (at 20h and
10h) correspond to their minima to the west. In order to show the diurnal
march of the declination in the northern hemisphere, in its generality and ana-
logy, I take the denominations employed by Sabine ; and using the mean local
time of each place, and beginning with the greatest westerly elongation we
have : —

Freiberg.

18-29.

Breslau.
1836.

Greenwich.
1846—47.

Makerstoun.
1842—43.

Toronto.
1845—47.

Washington.
1840-42.

Maximum

1

1

2

0 40'

1

2

Minimum

13

10

12

10

10

10

Maximum

16

16

16

14 14

14

14

Minimum

20

20

20

19 14

20

20

In particular seasons, Greenwich has shown some remarkable differences. In
1847 there was, in the winter, only one maximum (at 2h) and one minimum
(at 12h) ; in the summer there was a double progression, but the second mini-
mum occurred at 14h instead of at 16h (p. 236); the greatest westerly elonga-
tion remained attached to 2h in summer as well as in winter. In 1846 (p. 94) the
secondary minimum was, in summer, as usual, at 20b, but in winter at 12b.
The mean winter increase to the west continued in this year, without interrup-
tion, from midnight to 2h. Compare also 1845 (p. 5). Makerstoun (Rox-
burghshire in Scotland) is the observatory due to the munificent scientific zeal
of Sir Thomas Brisbane. (See J. A. Broun, Observations in Magnetism and
Meteorology, made at Makerstoun in 1843, p. 221 — 227.) On the hourly
day and night observations at St. Petersburg, see Kupffer, Compte rendu me'-
te'or. et magn. a M. de Brock en 1851, p. 17. Sabine points out, in reference
to his interesting and very sagaciously combined graphical representation of the
hourly declination curve of Toronto (Phil. Trans, for 1851, Pt. II. Plate 27), how,
previous to the small nocturnal movement to the west, which begins at llh and
lasts till 15h, there occurs a singular pause, lasting two hours, from 9h to llh.
" We find," Sabine remarks, " alternate progression and retrogression at Toronto
twice in the twenty-four hours. In two of the eight quarters (1841 and 1842)
the inferior degree of regularity during the night occasions the occurrence of a
triple max. and min.; in the remaining quarters the turning hours are the same
as those of the mean of the two years." (Obs. made at the Magn. and Meteor.
Observatory at Toronto in Canada, vol. i. p. xiv. xxiv. 183 — 191 and 228;
and Unusual Magn. Dist. Pt. I. p. vi.)

For the very complete observations at Washington, see Gilliss, Magn. and
Met. Observations made at Washington, p. 325 (General Law). Compare there-

NOTES. XXXV

•with Bache, Observations at the Magn. and Met. Observatory at the Girard Col-
lege, Philadelphia, made in the years 1840—1845 (3 vols., containing 3212
pages), vol. i. p. 709, vol. ii. p. 1285, vol. iii. p. 2167 and 2702. Notwith-
standing the vicinity of the two places (Philadelphia is only 1° 4' north, and
Qh jm 335 easfc Of Washington), I find a difference in the times of the westerly
secondary maximum and secondary minimum ; the former being an hour and a
half, and the latter two hours and a quarter, earlier in Philadelphia than in
Washington.

(155) p. 129. — I find instances of such small differences of time given by
Lieutenant Gilliss, in his Magnetic Observations at Washington, p. 328. In the
more northern latitude of Makerstoun in Scotland (lat. 55° 35'), there are fluc-
tuations in the secondary minimum, which in the four last and three first months
of the year takes place at 21h, but in the other five months (April to August)
at 19h; contrary, therefore, to Berlin and Greenwich. (Broun, Obs. made at
Makerstoun, p. 225.) The nocturnal movements of the needle, the secondary
maximum and secondary minimum, speak most clearly against the regular
diurnal variation of the declination being due to thermal variations; (the morning
and principal minimum does take place not far from the minimum of tempera-
ture, and the principal maximum near the maximum of temperature). " There
are two maxima and two minima of declination in the twenty-four hours, but only
one maximum and one minimum of temperature in the same period." (Relshuber,
in Poggend. Annalen der Physik und Chemie, Bd. 85, 1852, S. 416.) On
the normal march of the magnetic needle in North Germany, see an excellent
and faithful description in a Memoir of Dove's (Poggend. Ann. Bd. xix. S. 364 —
374).

(156) p. 129. — Voyage en Islande et au Greenland, exe'cute" en 1835 et en
1836, sarla corvette " la Recherche;" Physique (1838), p. 214 — 225 and
358—367.

(!57) p. 129. — Sabine, Account of Pendulum Experiments, 1825, p. 500.

(153) p. 130. — See Barlow's Account of the Observations at Port Bowen, in
the Edinb. New. Phil. Journal, vol. ii. 1827, p. 347.

(159) p. 130. — Professor Orlebar, at Oxford, at one time superintendent of the
Magnetic Observatory established at the expense of the East India Company
on the Island of Colaba, has tried to make out the complicated laws of the va-
_riation of the declination in the sub-periods (" Observations made at the Magn.
and Met. Observatory at Bombay, in 1845;" Results, p. 2 — 7). I have been
struck, in the first period, from April to August, by the agreement with the
march of the needle in Middle Europe (W.min. 19^h, max. 0£h, min. 5£h, mas.
7h). The month of October is a transition period, for in November and Decem-

c 2

XXXVI NOTES.

her the diurnal variation scarcely reaches two minutes. Although Bombay is still
8° from the magnetic equator, the regularity of the turning hours is already
hard to recognise. Wherever in nature different kinds of disturbing causes act
in recurring, but wholly or partially unknown periods, then, inasmuch as the
disturbances often either counteract or unequally reinforce each other, the regu-
lating laws long remain concealed from us.

(16°) p. 131.— See the proofs in my Examen crit. de 1'Hist. de la Ge'ogr. t.
iii. p. 34 — 37. The oldest statement of any particular amount of declination
was by Keutsungchy, a writer of the beginning of the twelfth century, and was
E. § S. (Klaproth's Lettre sur 1'Invention de k Boussole, p. 68.)

(161) p. 131. — On the ancient intercourse of the Chinese with Java, according
to the accounts of Fahian in the Fo-kue-ki, see Wilhelm v. Humboldt, " Ueber
die Kawi-Sprache," Bd. i. S. 16.

(162) p. 131.— Phil. Trans, for 1793, p> 340—349; for 1798, p. 397. The
result which Macdonald derives from his observations at Fort Marlborough
(situated above the town of Bencoolen, in Sumatra, 3° 47' S. lat.), according
to which the easterly elongation increases from 19h to 5h, does not seem
to me to be quite warranted. After the hour of noon, the first observation
ordinarily taken was either at 3h, 4h, or 5U ; and some detached observations,
occasionally made at other hours, render it probable that in Sumatra, as at Ho-
barton, the turning hour from east to west was as early as 2h. We have twenty-
three months of declination observed by Macdonald (from June 1794, to June
1796), and it appears from them that at all seasons the east declination increased
from 19^h to noon, by a continuous movement of the needle from W. to E. We do
not find in these observations any trace of the type of the phenomena of the
northern hemisphere (Toronto movements), which prevails from May to Septem-
ber at Singapore ; although Fort Marlborough is almost under the same meridian
as Singapore, and only 5° 4' of latitude from it; the Earth's equator, however,
passing between the two places.

(Ii53) p. 132.— Sabine, Magn. Obs. made at Hobarton, vol. i. (1841 and 1842)
p. xxxv. p. 2 and 148; vol. ii. (1843— 1845) p. iii.— xxxv. and 172—344.
Compare also Sabine, Obs. made at St. Helena; the same in the Phil. Trans, for
1847, Pt. I. p. 55, PI. IV. and Phil. Trans, for 1851, Pt. II. p. 636, PL XXVII.

(164) p. 133.— Kosmos, Bd. i. S. 190 (English edition, p. 172).

(165) p. 134.— Sabine, Observations made at the Magn. and Meteor. Observa-
tory at St. Helena in 1840—1845, vol. i. p. 30, and the same in the Phil.
Trans, for 1847, Pt. I. p. 51 — 56. PI. III. The regularity of the opposition in
the two parts of the year, May to September following the type of the middle
latitudes of the northern hemisphere, and October to February the type of the

NOTES. xxxvii

middle latitudes of the southern hemisphere, is seen in the graphical representa-
tion, with all its striking distinctness, when we compare severally the form and
inflexions of the curves of horary variation in the parts of the day from 14h to 22h,
from 22h to 4h, and from 4h to 14h. Each inflexion above the line of mean declina-
tion corresponds to an almost equal or similar one below it (vol. i. PI. IV. curves
AA and BB). Even during the night the opposition is sensible, and it is a very
interesting remark that when at St. Helena and the Cape of Good Hope the type
is that of the northern hemisphere, there is seen at these very southern places
the same anticipation of the turning hours as that which takes place in the same
months at Toronto in Canada. Sabine, Observations at Hobarton, vol. i. p. xxxvi.
(I6a) p. 135.— Phil. Trans, for 1847, Pt. I. p. 52 and 57, and Sabine, Observa-
tions made at the Magn. and Met. Observatory at the Cape of Good Hope, 1841 —
1846, vol. i. p. xii — xxiii. PI. III. (Compare also Faraday's ingenious views on
the causes of such phenomena in their dependence on the seasons of the year, in
his " Experiments on Atmospheric Magnetism," § 3027 — 3068, and on the ana-
logies with St. Petersburg, § 3017.) Near the southern shores of the Red Sea,
a very diligent observer, M. d'Abbadie, is believed to have observed the same re-
markable change of type in the march of the declination in the opposite parts of
the year as that found at the Cape, St. Helena, and Singapore. (Airy, On the
present State of the Science of Terrestrial Magnetism, 1850, p. 2.) Sabine re-
marks, in the Proceedings of the Royal Society, 1849, p. 821 : " It results, from
the present position of the four points of maximum force at the surface of the
Earth, that the intermediate line of least force departs considerably in the
Southern Atlantic from the middle or geographically equatorial portion of the
globe, passing between the Cape and St. Helena, and consequently not far from
either of those stations. The latitude of the Cape is 33° 56' S. It is conse-
quently not situated within the tropics, and the sun is throughout the year well
to the north of its zenith, and therefore, according to M. de la Rive's thermal
theory (Ann. de Chim. et de Phys. t. xxv. 1849, p. 310), the deviation should
be in one and the same direction throughout the year. But the fact is not so, for
the same contrariety in the direction of the diurnal variation takes place at the
Cape as at St. Helena ; the two portions of the year at which the opposite phae-
nomena prevail are also identical at the two stations; and at both the change in
the direction of the deviation takes place about the time when the sun crosses
the equator ; the deviation being to the west at both stations when the sun is
north of the equator, and to the east when south of the equator."

(167) p. 135. — Halley, "Account of the late surprising Appearance of Lights in
the Air," in the Phil. Trans, vol. xxix. 1714 — 1716, No. 347, p. 422 — 428.
Unfortunately, Halley's explanation of the Northern Light is connected with the

XXXV111 NOTES.

fanciful hypothesis propounded by him twenty-five years before (Phil. Trans, for
1693, vol. xvii. No. 195, p. 563), in which he supposed the Earth on which we
dwell to be a hollow shell, enclosing an interior smaller globe also supporting
human beings ; and that " in order to make that inner globe capable of being inha-
bited, there might not improbably be contained some luminous medium between
the balls, so as to make a perpetual day below." Then, supposing that the Earth's
compression near the poles of rotation must cause the outer shell to be thinner
there than at the equator, he considered that at certain seasons, and especially at
the equinoxes, the interior luminous and magnetic fluid seeks a vent through
fissures of the rocks composing the thinner crust near the poles ; and this
streaming forth of the bright fluid is, according to Halley, the phenomenon of the
aurora. Experiments with iron-filings strewed over a spheroidal-shaped magnet
(" a terrella"), served, he considered, to explain the direction of the luminous and
coloured rays of the aurora. " As every one has his own rainbow, so every ob-
server sees the corona at a different point." On the geognostic fancies of this
ingenious, and 'in all his magnetical and astronomical works, solid philosopher,
compare Kosmos, Bd. i. S. 178 and 425, Anm. 6 (English edition, p. 161 and
Note 136).

(I6S) p. 137. — In the great fatigue of observing for many successive nights,
Oltmanns and I were sometimes relieved or aided by very trustworthy observers —
the architect Herr Mampel, the geographer Herr Friesen, the highly informed
mechanician Nathan Mendelsohn, and our great geognost Leopold von Buch.
It is always a pleasure to me to name here, as hi all my earlier writings, those
who have kindly participated in my labours.

(le9) p. 138. — The month of September, 1806, was remarkably rich in mag-
netic storms. To exemplify this, I subjoin an extract from my journal :

§£ Sept. 1806 from 16 36 to 17 43

If „ „ 16 40 „ 19 2

\\ „ „ 15 33 „ 18 27

8 » » I* 4 „ 18 2

I „ » H 22 „ 16 30
1? „ „ 14 12 „ 16 3

II » „ 13 55 „ 17 27
i 12 3 13 22

The last was a small storm, and was followed by the greatest calm, which lasted
through the whole night and following morning until noon. On the 29th to 30th
Sept. 1806, at 10h 20m a small storm until llh 32m,and then great calm until
17h 6m.

NOTES. XXXIX

TSr 1806.— At 14h 46m, a great storm, but of short duration ; then en-
tire quiet ; and at 16h 30m another equally great storm.

The great storm of the f|th Sept. had been preceded by a still stronger one,
lasting from 7b 8m to 9h llm. During the winter which followed, the number of
disturbances was very small, and at no time to be compared with the autumnal
disturbances. I give the name of " great storm" to a state in which the needle
makes oscillations of from 20 to 38 minutes of arc, or passes altogether out of the
scale, or moves so rapidly as to make any observation impracticable. In " small
storms" the oscillations are irregular, and of from 5 to 8 minutes in amount.

(l:o) p. 138. — During ten years of diligent observation at Paris, up to 1829,
Arago never saw oscillations without their being accompanied by a " change of
declination." He wrote to me in the above-named year: "J'ai communique a
1'Acade'mie les resultats de nos observations simultane'es. J'ai e^e' surpris des
oscillations qu'eprouve parfois 1'aiguille de de'clinaison h, Berlin dans les observa-
tions de 1806, 1807, et de 1828 et 1829, lors m£me que la de'clinaison moyenne
n'est pas alte're'e. Ici (a Paris) nous ne trouvons jamais rien de semblable. Si
1'aiguille e'prouve de fortes oscillations, c'est seulement en terns d'aurore boreale,
et lorsque sa direction absolue a etc uotablement de'range'e ; et encore le plus
souvent les derangements dans la direction ne sont-ils pas accompagnes du
mouvement oscillatoire." It is quite otherwise, however, with the phaenoinena
observed at Toronto (N. lat. 43° 39'), which agree entirely with those at Berlin.
The observers at Toronto were very attentive to this kind of motion, and noted
" strong " and " slight vibrations," shocks, and all degrees of disturbance,
according to determinate subdivisions of a definite scale, and according to a
determinate and uniform nomenclature. (Sabine, Days of Unusual Magnetic
Disturbances, vol. i. Pt. I. p. 46.) In this volume, comprising the observations
of 1840 and 1841, an account is given in detail of six groups of successive days
(amounting in all to 146 days in those two years), in which the oscillations were
often very considerable (" with strong shocks "), without sensible change, from
the normal declination appropriate to the hour. Such groups are designated by
the heading, " Times of Observation at Toronto, at which the Magnetometers
were disturbed, but the mean readings were not materially changed." (See pp.
47, 54, 74, 88, 95 and 101.) The alterations in the declination during the
frequently occurring displays of aurora borealis were almost always accompanied
at Toronto by strong oscillation ; often such as even to make readings impossible.
We learn, then, by these phaenomena (the still further examination of which can-
not be too earnestly recommended), that although great momentary disquiet is
often followed by considerable definitive variation in the declination (Young-
husband, Unusual Disturbances, Pt. II. p. x.), yet that, on the whole, the magni-

xl NOTES.

tude of the arc of oscillation is by no means generally proportioned to the
magnitude of the alteration of the declination; that the oscillations may be very
great, while the alteration of the declination is quite inconsiderable; and that,
on the other hand, the progression of the needle in east or west declination may
be rapid and considerable, without any vibratory movement; and also that these
manifestations of magnetic activity may assume at different places a peculiar and
different character.

(Irl) p. 139.— Unusual Disturbances, vol. i. Pt. I. p. 69 and 101.

(172) p. 139. — This was at the end of September 1806. The fact was
published in Poggendorffs Annalen der Physik, Bd. xv. (April 1829) S. 330.
It is there said — " My older hourly observations, made in conjunction with
Oltmanns, had the advantage that at that time (1806 and 1807) no similar
ones had been made either in France or in England. They gave the nocturnal
maxima and minima [of the diurnal variation, — Ed.] ; and they made known the
remarkable magnetic storms, which, by the strength of the oscillatory movement
of the needle, often make observation impossible, and which return for several
successive nights about the same time, without its having yet been possible to
discern the operation of any meteorological relations." It was not, therefore, in
1839 that a certain periodicity [i. e. a disposition to return at the same hours
on successive nights, — Ed.] in the extraordinary disturbances was first recognised.
(Report of the Fifteenth Meeting of the British Association at Cambridge, 1845,
Pt. II. p. 12.)

(173) p. 139. — Kupffer, Voyage au Mont Elbruz dans le Caucase, 1829,
p. 108: " Les deviations irre'gulieres se re'petent sou vent a la m&ne heure et
pendant plusieurs jours conse'cutifs."

(m) p. 140. — Sabine, Unusual Disturb., vol. i. Pt. I. p. xxi., and Young-
husband, On periodical Laws in the larger Magnetic Disturbances, in the Phil.
Trans, for 1853, Pt. I. p. 173.

(17S) p. 140. — Sabine, in the Phil. Trans, for 1851, Pt. I. p. 125 to 127:
" The diurnal variation observed is, in fact, constituted by two variations super-
posed upon each other, having different laws, and bearing different proportions to
each other in different parts of the globe. At tropical stations the influence of
what have been hitherto called the irregular disturbances (magnetic storms) is
comparatively feeble; but it is otherwise at stations situated as are Toronto
(Canada) and Hobarton (Van Diemen Island), where their influence is both
really and proportionally greater, and amounts to a clearly recognisable part of
the whole diurnal variation." There takes place here, in the complex result of
different causes acting at the same time, that which Poisson has so well described
in the theory of waves (Annales de Chimie et de Physique, t. vii. 1817, p. 293):

NOTES. xli

" Plusieurs sortes d'ondes pcavent se croiser dans 1'eau comme dans Pair; les
petits mouvements se superposent" Compare Lament's conjectures on the
compound action of a polar and an equatorial wave, in Poggend. Annalen, Bd.
Ixxxiv. S. 583.

(176) p. 141.— See above, p. 138, Note 169.

(177) p. 142.— Sabine, in the Phil. Trans, for 1852, Pt. II. p. 110; Young-
husband, in the before-quoted work, p. 169.

(178) p. 142. — According to Lamont and Eelshuber, the magnetic period is
10^ years; so that the amount of the mean diurnal movement of the needle
would continually increase for five, and decrease for five years ; the winter move-
ment (or amplitude of the declination) being little more than half as great as
that of the summer months. (Compare Lamont, Jahresbericht der Sternwarte zu
Munchen fur 1852, S. 54—60.) The director of the Berne Observatory, Eudolph
Wolf, finds, by an enquiry extending over a much longer space of time, a con-
current period of the magnetic declination and of the frequency of the solar spots
of ITl years.

(lre) p. 142.— Kosmos, Bd. iv. S. 74, 75 (Anm. 73), 77, 80, and 81 (English
edition, p. 80, 81 (Note 73), 85, 88, and 89).

(I8e) p. 142. — Sabine, in the Phil. Trans, for 1852, Pt. I. p. 103 and 121.
Compare, besides the memoir of Rud. Wolfs, already referred to (present volume,
p. 81), of July 1852, and similar conjectures published, almost at the same time,
by Gautier, in the Bibliotheque Universelle de Geneve, t. xx. p. 189.

(181) p. 143.— Kosmos, Bd. iii. S. 401—403 (English edition, p. 291—293).

(I32) p. 143.— Sabine, in the Phil. Trans, for 1850, Pt. I. p. 216. (Faraday,
Exper. Researches on Electricity, 1851, p. 56, 73, and 76; § 2891, 2949, and
2958.)

183) p. 143. — Kosmos, Bd. i. S. 185 and 427, Anm. 13 (English edition,
p. 168 and Note 143); Poggend. Annalen, Bd. .xv. S. 334 and 335; Sa-
bine, Unusual Disturb, vol. i. Pt, I. p. xiv — xviii., where are given tables of
simultaneous storms at Toronto, Prague, and Hobarton. On days when the
magnetic storms were most violent in Canada (22nd March, 10th May, 6th
Aug., and 25th Sept., 1841), similar phenomena showed themselves in the
southern hemisphere. Compare also Sir Edward Belcher, in the Phil. Trans, for
1843, p. 133.

(18t) p. 144.— Kosmos, Bd. i. S. 219 (English edition, p. 199).

(18S) p. 145.— Kosmos, Bd. i. S. 188, 189, and 430, Anm. 20—22 (English
edition, p. 171 and Notes 150 — 152); Bd. ii. S. 319—321 and 482. Anm.
93 and 94 (English edition, p. 282 and 283, Notes 434 and 435); Bd. iv.
S. 51—60 (English edition, p. 52—62).

xlii NOTES.

(W6) p. 146. — At very different periods, face in 1809, in my Recueil d'Ob-
serv. astron. vol. i. p. 368, and again in 1839, in a letter to Lord Minto, then
First Lord of the British Admiralty, a few days after the departure of Sir James
Ross on the Antarctic Expedition, I have described more fully the importance
which I attach to the proposal alluded to in the text. (Compare Report of the
Committee of Physics and Meteor, of the Royal Soc. relative to the Antarctic
Exped. 1840, p. 88 — 91.) " Suivre les traces de 1'e'quateur magne'tique ou
celles des lignes sans de'clinaison, c'est gouverner (diriger la route du vaisseau)
de maniere a couper les lignes zero dans les intervalles les plus petits, en chan-
geant de rumb chaque fois que les observations d'inclinaison ou de de'clinaison
prouveht qu'on a de'vie*. Je n'ignore pas que d'apres de grandes vues sur les
ve'ritables fondements d'une Theorie generate du Magnetisme tewestre, dues a
M. Gauss, la connaissance approfondie de Yintensite horizontal, le choix des
points ou les trois elements de de'clinaison, d'inclinaison et d'intensite' totale ont

y

e'te* mesures simultanement, suffisent pour trouver la valeur de — (Gauss, § 4

B

»:t 27), et que ce sont la les points vitaux des recherches futures; mais la somme
des petites attractions locales, les besoins du pilotage, les corrections habituelles
du rumb et la securite des routes continuent k donner une importance speciale a
la connaissance de la position et des mouvements de translation peYiodique des
lignes sans dcclinaison. Je plaide ici leur cause, qui est lie'e aux inte'rSts de la
ge'ographie physique." Many years must pass away before declination-charts,
constructed from the theory of terrestrial magnetism, can be such as to afford
adequate guidance to navigation (Sabine, in the Phil. Trans, for 1849, Pt. II.
p. 204); and the entirely objective view which I here advocate, if it led to
periodically repeated determinations by sea and land expeditions undertaken for
the purpose, and made, therefore, at the same time, would afford the twofold
advantage of immediate practical use, and of an exact knowledge of the progres-
sive secular movement of the lines, as well as of obtaining a large supply of
fresh data as bases of calculation for the theory of terrestrial magnetism founded
by Gauss (Gauss, § 25). It would be highly expedient, for the sake of the
exact determination of the movement of the lines of no inclination and no decli-
nation, to establish substantial and conspicuous land-marks at the points where
these lines enter or quit the coasts of continents for the years 1850, 1875, 1900.

In such expeditions, similar to the ancient ones of Halley, many

other isoclinal and isogonic lines, besides those of no inclination and no declina-
tion, would be crossed and their position determined ; and the horizontal and
total magnetic force could also be measured (at least on shore); and thus several
objects would be attained at the same time. I find the wish which I have thus

NOTES. xliii

expressed supported by a great nautical authority to whom I always refer with
great pleasure, Sir James Ross (Voyage in the Southern and Antarctic Regions,
vol. i. p. 105).

(187) p. 146. — Acosta, Historia de las Indias, 1590, lib. i. cap. 17. I have
touched in a previous volume on the question whether, through the medium of
the controversy between Boud and Beckborrow, the opinion of Dutch navigators
respecting four lines of no declination may not have influenced Halley's hypo-
thesis of four magnetic poles. (Kosmos, Bd. ii. S. 483; English edition, Note
434.)

(188) p. 147. — In a general point of view, especial attention is due to the
isogonic line of 22\° W. in the interior of Africa: this line connects very diffe-
rent systems of isogonic lines and, according to Gauss's theoretically constructed
map, runs from the eastern part of the Indian Ocean across Africa and the
Atlantic to Newfoundland. Perhaps the honourable extension which the British
Government have given, in the present year, to the African Expedition of
Richardson, Barth, and Overweg, may lead to the solution of this magnetic
problem. [The particular year in which this note was written is not named.
The reference is probably to the employment of Dr. Vogel, which was specially
for magnetic determinations. — Ed.]

(189) p. 147. — Sir James Ross crossed the line of no declination in 61|° S.
lat. and 22° 28' W. long. (Voyage to the Southern Seas, vol. ii. p. 357.) In
70° 43' S. lat. and 16° 46' W. long., in March 1843, Captain Crozier found the
declination 1° 38', and was therefore very near the line of no declination.
Compare Sabine, On the Magnetic Declination in the Atlantic Ocean for 1840,
in the Phil. Trans, for 1849, Pt. II. p. 233.

(19°) p. 148. — Sir James Ross, Voyage to the Southern Seas, vol. i. p. 104,
310 and 317.

(191) p. 148.— Elliot, in the Phil. Trans, for 1851, Pt. I. p. 331, PI. XIII.
" Sandal-wood Island " is a small island on which sandal-wood (in Malay and Ja-
vanese, "tschendana;" Sanscrit, " tschandana;" Arabic, "fsandel") is obtained.

(192) p. 149. — According to Barlow, and according to the map entitled " Lines
of Magnetic Declination computed according to the Theory of M. Gauss," in
the Report of the Committee for the Antarctic Expedition, 1840. According to
Barlow, the line of no declination coming from Australia enters the Asiatic con-
tinent at the Gulf of Cambay and then turns again immediately north-eastward,
by Thibet, China, and Formosa, to the Sea of Japan. According to Gauss, the
Australian line ascends simply through Persia, by Nishnei Novgorod, to Lap-
land. This great geometer regards the line of no declination of the seas of Japan
and the Philippines, as well as the closed oval group of isogonic lines in Eastern

xllV NOTES.

Asia, as quite unconnected with the line of no declination of Australia, the Indian
Sea, Western Asia, and Lapland.

(193) p. 149. — I have spoken elsewhere (Asie Centrale, t. iii. p. 458 — 461)
of this identity, founded on my own observations of the declination on the Cas-
pian, at Uralsk on the Jaik River, and in the steppe near Lake Elton.

(194) p. 149.— Adolf Erman's Map of the Magnetic Declination, 1827 — 1830.
Elliot's Map shows decidedly that the Australian line of no declination does not
traverse Java, but runs parallel to the south coast of that island, at a distance of
a degree and a half of latitude to the south of it. According to Erman (not
according to Gauss), the Australian line of no declination passes between Ma-
lacca and Borneo, through the Sea of Japan, to the closed oval group of Eastern
Asia, entering the continent at the north shore of Ochotsk (in lat. 59^°), and
yet redescends through Malacca, so that the ascending and descending portions of
the line are there only eleven degrees apart; and in this representation the west
Asiatic line of no declination (from the Caspian to Russian Lapland) is an im-
mediate continuation of the portion which comes down from north to south.

(195) p. 150.— In 1843, in my Asie Centrale (t. iii. p. 469—476), I remarked,
from documents preserved in the archives at Moscow and at Hanover, that Leib-
nitz (who had proposed the first plan for a French expedition to Egypt) had
also been the first who had sought, by availing himself of the relations established
in Germany, in 1712, with Czar Peter the Great, to get the position of the lines
of declination and inclination determined throughout the Russian Empire — an
extent exceeding that part of the moon's surface which is seen from the Earth —
and to arrange that this determination should be repeated at given epochs. In
a letter written to the Czar, which Pertz has discovered, Leibnitz refers to a
small hand-globe (terrella), still preserved at Hanover, on which he had drawn
the curve for which the declination is 0, his " linea magnetica primaria." He
maintained that there is only one line of no declination, dividing the globe into
two nearly equal parts, having four " puncta flexus contrarii," or sinuosities,
running from Cape de Verde towards the American coast in 36°, and then
directing itself through the Pacific towards the east of Asia and New Holland.
He made this line pass not through the poles of the Earth, but nearer to the
South Pole, where he supposed the declination to be 5°, than to the North, where
he supposed it to be 25°. He further considered the movement of this important
curve to be, in the beginning of the 18th century, towards the north ; and east
declination, from 0° to 15°, to prevail over great part of the Atlantic, through-
out the Pacific, in Japan, and in part of China and of New Holland. He pro-
posed that, the surgeon Donelli being dead, he should be succeeded by another.
" who might give few medicines but many good scientific counsels for making the

NOTES. xv

determinations of declination and inclination." .... This hitherto unregarded
document of Leibnitz does not manifest any particular theoretical views.

(196) p. 150. — See my magnetic observations in "Asie Centrale," t. iii. p. 400.

(197) p. 150. — Erman, Astron. und magn. Beobachtungen (Reise urn die
Erde, Abth. II. Bd. ii.), S. 532.

(19S) p. 150.— Hansteen, in Poggend. Ann. Bd. xxi. S. 371.

(199) p. 152. — Sabine, Magn. and Meteor. Observ. at the Cape of Good Hope,
vol. i. p. Ix.

(20°) p. 152. — In judging of such near epochs, we must not forget, in reference
to this subject, how easily, with the instruments and methods then in use, an
«rror of a degree might occur.

C201) p. 153. — Kosmos, Bd. i. S. 430, Anm. 20 (English edition, Note
150).

(?02) p. 153.— Euler, in the Mem. de 1'Acad. de Berlin, 1757, p. 176.

(203) p. 153.— Barlow, in the Phil. Trans, for 1833, Pt. II. p. 671. Great
uncertainty prevails as to the older magnetic observations at St. Petersburg, in
the first half of the 18th century. They would make the declination to have
been always 3° 15', or 3° 30', for the whole period fr9m 1726 to 1772! (Han-
steen, Magnetismus der Erde, S. 7 and 143.)

(204) p. 154.— Kosmos, Bd. i. S. 198—210 (English edition, p. 179—189) ;
and Dove, in Poggend. Ann. Bd. xix. S. 388.

(203) p. 154. — The meritorious examinations of Lottin, Bravais, Lilliehb'b'k,
and Siljestrom, who observed the phenomena of auroras from September 19,
1838, to April 8, 1839, at Bossekop in Finnmarken (lat. 69° 58'), and at
Jupvig (lat. 70° 6'), have been published in Part IV. of the " Voyages en
Scandinavie, en Laponie, au Spitzberg, et aux Feroe, sur la corvette 'la Recherche '
(aurores boreales)." To these observations are added important results obtained,
from 1837 to 1840, by English superintendents employed in the copper mines
at Kalfiord (lat. 69° 56'), p. 401—435.

(206) p. 154. — Compare p. 437 — 444 of the above-named work, on the subject
of the " Segment obscur de 1'Aurore bore'ale."

(207) p. 154. — Schweigger's Jahrbuch der Chimie und Physik, 1826, Bd. xvi.
S. 198, and Bd. xviii. S. 364. The dark segment and the incontestable shooting
upwards of black rays or streamers, in which the luminous process is nullified
(? by interference), may recall to us Quet's "Recherches sur rElectrico-chhnie
dans le Vide," and Ruhmkorff's fine experiments in which, in spaces containing
only very rarefied air, red light streamed from the positive, and violet from the
negative, metallic ball, but the intensely bright parallel strata of beams were
regularly separated by wholly dark strata: "La lumiere repandue entre les

xlvi NOTES.

boules terminates des deux conducteurs e'lectriques se partage en tranches nom«
breuses et paralleles, se'parees par des couches obscures alternantes, et rdguliere-
ment distinctes." (Comptes rendus de 1'Acad. des Sciences, t. xxxv. 1852,
p. 949.)

(20S) p. 155. — Voyages en Scandinavie (aurores bor.), p. 558. On auroral
coronas and canopies, see the excellent Accounts and Inquiries by Bravais, p.
502—514.

(209) p. 155. — Same work (draperie ondulante, flamme d'un navire de guerre
cleploye'e horizontalement et agLtee par le vent, crochets, fragments d'arcs et de
guirlandes), p. 35, 37, 45, 67, and 481. Mr. Bevalet, the distinguished artist
of the expedition, has furnished an interesting collection of drawings of the
various forms.

(21°) p. 155. — Compare Voy. en Scand. (aur. bor.), p. 523 to 528 and 557.

(2n) p. 156. — Kosmos, Bd. i. S. 201 and 441, Anm. 44 (English edition,
p. 182, 183, and Note 174). Compare Franklin, Narrative of a Journey to
the Shores of the Polar Sea in 1819 — 1822, p. 597; and Kamtz, Lehrbuch der
Meteorologie, Bd. iii. (1836) S. 488—490. The oldest conjectures respecting
a connection between the aurora and the formation of clouds, are probably those
of Frobesius. (See Auroras Borealis Spectacula, Helmst. 1739, p. 139.)

(2I2) p. 156. — I take a single example from my manuscript journal in my
Siberian journey. "The night of the |th August (1829) was passed by me
in the open air, separated from my travelling companions, at the Cossack post
of Krasnaia larki, the most eastern one on the Irtysch, and on the boundary of
Chinese Dzungarei, and therefore of some importance as regards astronomical
determination of latitude and longitude. The night was extremely clear. Before
midnight, there suddenly formed, on the eastern part of the heavens, polar bands
of cirrus (' de petits moutons e'galement espace's, distribues en bandes paralleles
et polaires '). Greatest altitude 35°; the northern point of convergence moving
slowly towards the east. They disappeared without reaching the zenith, and a
few minutes afterwards similar bands of cirrus appeared on the north-eastern
part of the sky. During the greater part of the remainder of the night, almost
until sunrise, they again moved in a very regular manner to N. 70° E. There
were an unusual number of shooting stars in the course of the night, and coloured
halos round the moon. There were no traces of aurora proper. A little rain
fell with feathery clouds ; in the forenoon of the 6th the sky was again clear,
and fresh polar bands were formed, and remained without moving or altering
their azimuth, from N.N.E. to S.S.W., as I had often seen in Quito and in
Mexico." (The magnetic declination is easterly.)

(2ls) p. 156. — Bravais, who, contrary to my experience, almost always found

NOTES. xlvii

the direction of the cirrus bands at Bosekop, like the auroral arch, perpendicular
to the magnetic meridian (Voyage en Scandinavie; Phenomena de translation
dans les pieds de 1'Arc des Auroras bore'ales, p. 534 — 537), describes, with his
usual exactness, the shiftings of position of the true auroral arch, p. 27, 92, 122,
and 487. In the southern hemisphere, Sir James Ross also observed similar pro-
gressive changes of direction of the aurora australis, changing from W.N.W. —
E.S.E. to N.N.E. — S.S.W. (Voyage in the Southern and Antarctic Regions,
vol. i. p. 311.) The absence of colour seems to be very frequent in the aurora
australis (vol. i. p. 266; vol. ii. p. 209). On nights without aurora in Lap-
land, see Bravais, work before quoted, p. 545.

(2H) p. 157. — Kosmos, Bd. i. S. 440, Anm. 43 (English edition, Note
173). Auroral arches seen in daylight remind us of the strength of the
light in the nuclei and tails of the comets of 1843 and 1847, which were
seen in North America, at Parma, and London near, the sun. Kosmos, Bd. i.
S. 390, Anm. 13 (English edition, Note 43); Bd. iii. S. 563 (English edition,
p. 400).

(215) p. 157. — Comptes rendus de 1'Acad. des Sciences, t. iv. 1837, p. 589.

(216) p. 157. — Voyages en Scandinavie, en Laponie, &c. (aurores bore'ales),
p. 559; and Martins, Trad, de la Mete'orol. de Ksemtz, p. 460. On the sup-
posed height of the aurora, see Bravais, work above quoted, p. 549 and 559.

(217) p. 158.— Same work, p. 462.

(218) p. 158. — Sabine, Unusual Magnet. Disturbances, Pt. I. p. xviii. xxii.
3 and 54.

(219) p. 158.— Dove, in Poggend. Ann. Bd. xx. S. 333 to 341. The unequal
effect exercised by an aurora on the declination needle at places situated in very
different meridians, may, in many cases, lead to the determination of the place of
the acting cause, for the outbreak of the luminous magnetic storm is by no
means always to be sought at the magnetic pole itself; and, as Argelander had
before said, and Bravais has confirmed, the summit of the luminous arch some-
times deviates more than 11° from the magnetic meridian.

C220) p. 159. — " On the 20th December, 1806. Sky azure- blue, without
trace of cloud. Towards ten o'clock, there appeared in the N.N.W. a reddish-
yellow luminous arch, through which I could distinguish, with the telescope,
stars of the 7th magnitude; a lyrse, being almost directly under the highest
point of the arch, enabled me to determine the azimuth of that point. It was
situated rather to the west of the vertical plane of the magnetic declination.
The aurora, which was in the N.N.W., repelled the North pole of the needle; for
instead of progressing towards the west, as did the azimuth of the arch, it
returned towards the east. The changes of magnetic declination, which amount

xlviii NOTES.

usually at night in this month to from 2' 27" to S', increased during the course
of the aurora, progressively, and without any considerable oscillations, to a
difference of 26' 28". The deflection was least at 9h 12m, when the auroral
light was most intense. In the horizontal magnetic force, we found during the
aurora 97"73 for 21 vibrations, and at 21h 50m — therefore long after the
termination of the aurora, which had ended at 14h 10m — 97*'17 for the same
number of vibrations. The temperature of the room in which the vibrations of
the small needle were observed was, in the first instance, 3°'2 Cent., and in the
second, 2°'8. The intensity had therefore diminished a little during the aurora.
There were no coloured rings round the moon." (Extract from my magnetic
journal.) Compare Hansteen, S. 459.

(221) p. 159. — Sabine, On Days of Unusual Magn. Disturbances, Pt. I. p. xviii.
"M. Bravais conclut des observations de Laponie qne 1'intensite horizontale
diminue pendant la pe'riode la plus active du phenomene de 1'aurore bore'ale."
(Martins, p. 461.)

(-•") p. 1 59. — Delesse, " Sur 1'Association des Mine'raux dans les Roches qui
ont un pouvoir magne'tique e'leve'," in the Comptes rendus de 1'Acad. des So.
t. xxxi. 1850, p. 806; and Annales des Mines 4e se'rie, t. xv. (1849) p. 130.

(223) p. 159.— Reich, on Mountain and Rock Magnetism in Poggend. Ann.
Bd. Ixxvii. S. 35.

(K4) p. 160.— When, in 1796, I filled the post of Superintendent of Mines in
the Fichtelgebirge in Franconia, and discovered the effects of the remarkable
Serpentine Mount (the Haidberg), which at some places affects the declination
of needles 24 feet distant, this question was raised. (Intelligenz-Blatt der all-
gem. Jenaer Litteratur-Zeitung, Dec. 1796, No. 169, S. 1447, and March
1797, No. 38, S. 323—326; Gren's Neues Journal der Physik, Bd. iv. 1797,
S. 136; and Annales de Chimie, t. xxii. p. 47.) I thought I had found that
the magnetic axes of the mountain were directly inverted in relation to those of
the Earth; but, according to the investigations of Bischoff and Goldfuss (Beschrei-
bung des Fichtelgebirges, Bd. i. S. 196), there were indeed recognised in 1816
magnetic axes traversing the Haidberg, and presenting opposite poles at the op-
posite declivities, but yet the azimuthal direction of the axes differed from that
which I had assigned. The Haidberg consists of green serpentine rock, partly
passing into choride-slate and hornblende-slate. In South America we found at
the village of Voysaco, in the Cordillera of Pasto, dykes of clay porphyry, and,
in the ascent of Chimborazo, groups of columnar trachyte, which affected the
needle at three feet distance. I was struck by finding in the black and red obsi-
dians of the Quinche, north of Quito, as well as in the grey obsidian of the Cerro
de las Navajas, of Mexico, large fragments having decided poles. All the con-

NOTES. xlix

magnetic mountains of the Oural chain, as well as that of Blago-
dat at Kuschwa, Wyssokaja Gora near Nishne Tagilsk, and Katschkanar at
Nishne Turinsk, rise out of and have broken through augitic, or rather oura-
litic porphyry. In the magnetic mountain of Blagodat, which I visited with
Gustave Rose in the Siberian Expedition of 1829, it does not appear that the
general action of the parts, which taken separately are found to have magnetic
poles, has produced any determinate and recognisable general magnetic axes.
Opposite poles lie near each other, irregularly intermingled. The same had pre-
viously been found by Erman (Reise um die Erde, Bd. i. S. 362). On the
degree of strength of polar force in serpentine, basaltic, and trachytic rocks, com-
pared to the quantity of interspersed particles of magnetic iron and oxyde of iron
in these rocks, as well as on the influence of atmospheric contact in developing
polarity, an influence spoken of earlier by Gme'lin and Gibbs, see the numerous
and very estimable observations of Zaddach, in his Beobachtungen iiber die mag-
' netische Polaritat des Basaltes und der trachytischen Gesteine, 1851, S. 56, 65
—78 and 95. By the comparison, in respect to polarity, of many basaltic frag-
ments of columns which had long stood exposed to the atmosphere, and of the
sides of other columns for the first time exposed to its contact, and by examining
other detached masses, taking away in doing so the earth which covered them,
first from their upper and then from their lower portions, Dr. Zaddach thought
he could infer (S. 74 and 80) that the polar property, which always appears to
be most strong with free access to the atmosphere, and in rock traversed by open
fissures, " spreads from the exterior towards the interior, and usually from above
downwards." Gme'lin said of the great magnetic mountain, Ulu-utasse-Tau, in
the country of the Bashkirs, near the Jaik: " The sides which are exposed to the
open air and daylight have the strongest magnetic force; those which are in the
earth are much weaker." (Reise durch Sibirien, 1740 — 1743, Bd. iv. S. 345.)
My great teacher, Werner, in his lectures on the Swedish magnetic iron, also
expressed his opinion " of the influence of atmospheric contact (not acting by
means of increased oxydation) in increasing polarity and attraction." In the mag-
netic iron mine of Succassung, in New Jersey, Colonel Gibbs states that " the ore
raised from the bottom of the mine has no magnetism at first, but acquires it
after it has been some time exposed to the influence of the atmosphere." (On
the Connexion of Magnetism and Light, in Silliman's American Journal of
Science, vol. i. 1819, p. 89.) Such a statement ought to cause exact experi-
ments to be undertaken. In remarking in the text (p. 160) that it is not only the
quantity of interspersed particles of iron in a particular kind of rock, but also their
relative distribution and arrangement, which affects the strength of the resulting
polar force, I considered those particles as so many small magnets. Compare
VOL. IV. d

] NOTES.

new views on this subject expressed by the great physicist Melloni, in a Memoir
read by him to the Royal Academy of Naples, in January 1853. (Esperienze
intorno al Magnetismo delle Rocche; Mem. I. sulla Polarita.) The old prejudice
which has prevailed so extensively, particularly in the Mediterranean; that by
rubbing a compass-needle with onions, or even by its being exposed to the exha-
lations of vinegar in which onions have been steeped, its directive force is less-
ened, and confusion in steering caused, is found mentioned in Procli Diadochi
Paraphrasis Ptolem. libri iv. de Siderum Affectionibus, 1635, p. 30. (Delambre,
Hist, de 1'Astronomie ancienne, t. ii. p. 545.) It is difficult to guess what may
have given occasion to this curious popular notion.

C225) p. 162.— Kosmos, Bd. iii. S. 44 (English edition, p. 35).

(226) p. 162.— Kosmos, Bd. i. S. 208—210 (English edition, p. 189—191).

(827) p. 163.— Kosmos, Bd. iii. S. 48, 431, 503, and 508—510 (English
edition, p. 39, 309, 360, 365—367.)

(S28) p. 164.— Kosmos, Bd. i. S. 220 (English edition, p. 200—201).

(z29) p. 165. — Kosmos, Bd. i. S. 233 (English edition, p. 212). Compare
Bertrand-Geslin, Sur les Roches lance'es par le Volcan de Boue du Monte Zibio
pres du Bourg de Sassuolo, in Humboldt, Voyage aux Re'gions dquinoxiales du
Nouveau Continent (Relation historique), t. iii. p. 566.

(23°) p. 165. — Robert Mallet, in the Transactions of the Royal Irish Academy,
vol. xxi. (1848) p. 51 — 113 ; the same, First Report on the Facts of Earth-
quake Phenomena, in the Report of the Meeting of the British Association for
the Advancement of Science, held in 1850, p. 1 — 89 ; the same, in the Manual
of Scientific Inquiry for the Use of the British Navy, 1849, p. 196—223 ;
William Hopkins, "On the Geological Theories of Elevation and Earthquakes," in
the Report of the Brit. Assoc. for 1847, p. 33—92. I have availed myself in
several particulars of the strict criticism to which Mr. Mallet has submitted my
earlier works in his very valuable memoirs (Irish Trans, p. 99 — 101, and Report
of Meeting of the Brit. Assoc. held at Edinburgh, p. 209).

(231)p. 165.— Thomas Young, Lectures on Natural Philosophy, 1807, vol. i.
p. 717.

(232) p. 166. — I follow the statistical statement communicated to me in 1802
by the Corregidor of Tacunga. It carried the loss to the number of from 30,000
to 34,000, but some twenty years later the number of those directly killed was
stated at a third less.

(833) p. 167.— Kosmos, Bd. i. S. 221 (English edition, p. 201).

(234) p. 169. — On this subject, and opposed to the supposition of molten " sub-
jacent fluid, confined in internal lakes" see Hopkins, Meeting of the Brit. Assoc.
in 1847, p. 57 j and on " the subterraneous lava tidal wave moving the solid

NOTES. H

crust above it," see Mallet, Meeting of Brit. Assoc. 1850, p. 20. Poisson also,
•with whom I more than once conversed on the hypothesis of subterranean tides
caused by the moon and sun, regarded the impulse, which he did not deny, as
inconsiderable, "as in the open sea the effect hardly amounts to 15 inches." On
the other hand, Ampere said, " Ceux qui admettent la liquidity du noyau inte-
rieur de la terre paraissent ne pas avoir songe' assez a 1'action qu'exercerait la
lune sur cette enorme masse liquide, action d'ou re'sulteraient des mare'es ana-
logues k celles de nos mers, mais bien autrement terribles, tant par leur e'tendue
que par la densite' du liquide. II est difficile de concevoir, comment 1'enveloppe
de la terre pourrait resister, e"taut incessamment battue par une espece de belief
hydraulique (?) de 1400 lieues de longueur." (Ampere, The'orie de la Terre
in the Revue des deux Mondes, juillet, 1833, p. 148.) If the interior of the
Earth is fluid, as in general we cannot doubt is the case, the particles still re-
maining movable, notwithstanding the enormous pressure, then there will be in
the interior the same conditions which, on the surface, produce the tides of the
ocean, and the tide-causing force will become weaker in approaching the centre,
as the difference of distance, from every two opposite points, regarded relatively
to the attracting heavenly bodies, will become less the greater the depth below
the surface, — and the force depends solely on the difference of the distances. If
the solid crust opposes resistance, the interior will at these places only exert a
pressure against the crust, but (as my astronomical friend, Dr. Brunnow, expresses
it) there will be no more tide than there would be if the ocean had an icy covering
which could not be burst open. The thickness of this solid unmolten crust has
been calculated from the melting point of different kinds of rock, and from the
increase of heat from the surface of the Earth downwards. I have already
(Kosmos, Bd. i. S. 27 and 48; English edition, p. 27 and Note 13) adduced
reasons in support of the conjecture that at about 201 geographical miles
below the surface the heat would attain the melting point of granite. Elie de
Beaumont (Geology, edited by Vogt, 1846, vol. i. p. 32) had made the very
similar estimate of 45,000 metres for the thickness of the solid crust. The in-
genious experiments of Bischof on the melting of different minerals, so important
to the progress of geology, would also lead to the assignment of a thickness of
between 115,000 and 128,000 French feet, or a mean of about 2l£ geographical
miles, for the thickness of the unmolten strata. See Bischof, Warmelehre des
Innern unsers Erdkorpers, S. 286 and 271. I am therefore the more surprised
to find that, assuming a definite limit between the solid and the molten materials
(not a gradual transition), Mr. Hopkins derives from the principles of his " Specu-
lative Geology " the result, that " the thickness of the solid shell cannot be less
than about one fourth or one fifth (?) of the radius of its external surface."

d 2

Hi NOTES.

(Meeting of the Brit. Assoc. held at Oxford in 1847, p. 51.) Cordier's earliest
assumption had only been 56 geographical miles, without correction for the in-
creasing pressure of the strata with increasing depth, and for the hypsometric
form of the surface. The thickness of the solid portion of the Earth's crust is
probably very unequal.

(2s5) p. 170.— Gay-Lussac, Reflexions sur les Volcans, in the Annales de
Chimie et de Physique, t. xxii. 1823, p. 418 and 426. Gay-Lussac, who,
together with Leopold von Buch and myself, had observed the great eruption of
Vesuvius in Sept. 1805, has the merit of having subjected the chemical hypo-
thesis to a strict critical examination. He sought the cause of volcanic pheno-
mena in an " affinite* tres-energique et non encore satisfaite entre les substances,
a laquelle un contact fortuit leur permettait d'obeir." His criticism was on the
whole favourable to the hypothesis of Davy and Ampere, " en supposant que les
radicaux de la silice, de 1'alumine, de la chaux, et du fer soient unis au chlore
dans 1'inte'rieur de la terre :" he also deemed it not improbable that, under certain
conditions, sea-water may penetrate (p. 419, 420, 423, and 426). On the diffi-
culties of a theory which rests on the penetration of water, compare Hopkins, in
the Meeting of the British Association, 1847, p. 38.

(238) p. 170. — Hydrochloric acid is entirely wanting in the vapours emitted
by the South-American volcanoes (according to the fine analyses of Boussingault
at five craters, viz. those of Tolima, Purace, Pasto, Tuqueras, and Cumbal), but
not in the Italian ones. (Annales de Chimie, t. lii. 1833, p. 7 and 23.)

(2s7) p. 170. •— Kosmos, Bd. i. S. 247 (English edition, p. 226). Davy,
while most distinctly giving up the opinion that volcanic eruptions are a conse-
quence of the contact of the metallic bases with air and water, yet stated his
belief that the existence of oxydisable metalloids in the interior of the Earth
might act as a concurrent cause in volcanic processes which had already begun.

C238) p. 170. — " Jattribue," said Boussingault, "la plupart des tremble-
ments de terre dans la Cordillere des Andes a des eTxmlements qui ont lieu dans
1'inte'rieur de ces montagnes par 1'entassement qui s'opere et qui est une conse-
quence de leur soulevement. Le massif qui constitue ces cimes gigantesques
n'a pas e'te souleve' a 1'dtat pateux; le soulevement n'a eu lieu qu'apres la solidi-
fication des roches. J'admets par consequent que le relief des Andes se compose
de fragments de toutes dimensions, entasse's les uns sur les autres. La consoli-
dation des fragments n'a pu etre tellement stable des le principe qu'il n'y ait des
tassements apres le soulevement, qu'il n'y ait des mouvements inte'rieurs dans ies
masses fragmentaires." (Boussingault, Sur les Tremblements de Terre des Andes,
In the Annales de Chimie et de Physique, t. Iviii. 1835, p. 84—86.) In the
description of his memorable ascent of the Chimborazo (Ascension au Chimborazo/

NOTES. liii

le 16 dec. 1831, above-quoted work, p. 176), he says again: — " Comme le
Cotopaxi, 1'Antisana, le Tunguragua et en general les volcans qui heVissent les
plateaux des Andes, la masse du Chimborazo est forme'e par 1'accumulation de
debris trachytiques, ainoncele's sans aucun ordre. Ces fragments, d'un volume
souvent e'norme, out e"te souleve's a 1'e'tat solide par des fluides e'lastiques qui se
sont fait jour sur les points de moindre resistance; leurs angles sont toujours
tranchants." The cause of earthquakes here indicated is that which Hopkins, in
his " Analytical Theory of Volcanic Phenomena," terms " a shock produced by
the falling of the roof of a subterranean cavity." (Meeting of the British Assoc.
at Oxford, 1847, p. 82.)

(™9) p. 171. — Mallet, Dynamics of Earthquakes, p. 74—82. All that we
know of the waves of shocks and vibrations in solid bodies, shows how untenable
are the older theories of the propagation of the motion being facilitated by a
succession of cavities. Cavities can only act in respect to earthquakes in an
indirect or secondary manner, in affording spaces for the accumulation of vapours
and condensed gases. Gay-Lussac says finely : — "La terre, vieille de tant de
siecles, conserve encore une force intestine, qui eleve des montagnes (dans la
crotite oxyde'e), renverse des cites et agite la masse entiere. La plupart des
moutagnes, en sortant du sein de la terre, ont du y laisser de vastes cavites qui
Bout resides vides, a moins qu'elles n'aient etc remplies par 1'eau (et des fluides
gaseux). C'est bien a tort que Delue et beaucoup de geologues se serrent de
ces vides, qu'ils imaginent se prolonger en longues galeries, pour propager au loin
les tremblements de terre. Ces phe'nomenes si grands et si terribles sont de tres-
f'ortes ondes sonores, excite'es dans la masse solide de la terre par une commotion
quelconque, qui s'y propage avec la meuie vitesse que le son s'y propagerait. Le
mouvement d'une voiture sur le pave ebranle les plus vastes e'din'ces, et se com-
munique a travers des masses conside'rables, comme dans les carrieres profondes
au-dessous de Paris." (Ann. de Chim. et de Phys. t. xxii. 1823, p. 428.)

(21°) p. 171. — On interference-phenomena in earthquake waves, analogous
to waves of sound, see Kosmos, Bd. i. S. 21 1 ; English edition, p. 191 — 192, and
Kleinere Schriften, Bd. i. S. 379.

(24:) p. 171. — Mallet, On Vorticose Shocks and Cases of Twisting, Meeting of
Brit. Assoc. 1850, p. 33 and 49; and in the Admiralty Manual, 1849, p. 213.
Compare Kosmos, Bd. i. S. 212 (English edition, p. 192.)

(24-) p. 171. — The Moya cones were seen by Boussingault nineteen years
after I had seen them. " Des eruptions boueuses, suites du tremblement de
terre, comme les eruptions de la Moya de Pelileo qui ont enseveli des villages
entiers." (Ann. de Chimie et de Phys. t. Iviii. p. 81.)

(243) p. 173. — On the removal of buildings and plantations in the earthquake

liv NOTES.

of Calabria, see Lyell, Principles of Geology, vol. i. p. 484—491. On escape
with life in fissures in the great earthquake of Riobamba, see my Relation hist.
t. ii. p. 642. As a remarkable instance of the closing of a fissure, it is related
by Scacchi that, in the earthquake which took place in the summer of 1851 in
the Neapolitan province Basilicata, at Barile near Melfi, a hen was found caught
by her two feet in the pavement of the street.

(244) p. 174. — Kosmos, Bd. i. S. 212 (English edition, p. 192). Hopkins
has shown theoretically, with great justness, how fissures produced in earth-
quakes may prove exceedingly instructive in reference to the formation of dykes
and the phenomena of faults, inasmuch as tbe later formed displaces the older
formation. Long, however, before the valuable labours of Phillips, Werner, in his
Theory of Veins (1791), had shown the relations of precedence in age between
the displacing traversing dyke and that which is displaced and traversed: com-
pare Report of Brit. Assoc. Meeting at Oxford, 1847, p. 62.

(245) p. 175. — On the simultaneous agitation of the tertiary calcareous strata
at Cumana and Maniquarez since the great earthquake of Cumana, 14th De-
cember, 1796, compare Humboldt, Relation hist. t. i. p. 314; Kosmos, Bd. i.
S. 220 (English edition, p. 200); and Mallet, Meeting of the Brit. Assoc., 1850,
p. 28.

(246) p. 175. — Abich on Daghestan, Schagdagh, and Ghilan, in PoggendorfTs
Ann. Bd. Ixxvi. 1849, S. 157. In consequence of the wide-spreading earth-
quake of 29th July, 1846, the centre of commotion of which is supposed to have
been about St. Goar on the Rhine, the salt-spring, in a boring at Sassendorff
in Westphalia (Regier. Bezirk Arnsberg), was found, by careful analysis, to have
increased its saline contents 1| per cent., probably through the opening of fresh
channels of influx (Noggerath, das Erdbeben im Rheingebiete, vom 29. Juli,
1846, S. 14). In the Swiss earthquake of the 25th of August, 1851, Charpen-
tier has remarked that the temperature of tbe sulphurous spring of Lavey (above
St. Maurice on the Rhone) rose from 31° to 36°'3 cent.

(247) p. 176. — At Schemacha, — at an elevation of 2245 French feet, one of
the many meteorological stations which were established under Abich's direction
by Prince Woronzow in the Caucasus, — eighteen earthquakes have been recorded
by the observer in the year 1848 only.

(M) p. 176.— See my Asie Centrale, t. i.pp. 324—329, and t. ii. pp. 108 —
120; and in particular my Carte des Montagnes et Volcans de 1'Asie, compared
with the geological maps of the Caucasus and of the highlands of Armenia, by
Abich, as well as with the Map of Asia Minor, by Pierre Tschichatschef, 1 853
(Rose, Reise nach dem Ural, Altai und Kasp. Meere, Bd. ii. S. 576 und 597). " Du
Tourfan, situe sur la pente me'ridionale du Thian-chan, jusqu'a 1'Archipel des

NOTES. lv

Azores, il y a 120° de longitude. C'est vraisemblablement la Tmnde de reactions
vokaniques la plus longue et la plus re'guliere, oscillant faiblement entre 38° et
40° de latitude, qui existe sur la terre ; elle surpasse de beaucoup en e'tendue la
bande volcanique de la Cordillere des Andes dans 1'Ame'rique me'ridionale. J'in-
siate d'autant plus sur ce singulier alignement d'ar^tes, de soulevements, de cre-
vasses, et de propagations de commotions, qui comprend un tiers de la circonfe-
rence d'un parallele a Vequateur, que de petits accidents de la surface, Fine'gale
hauteur et la largeur des rides ou soulevements line'aires, comme 1'interruption
cause'e par les bassins des mers (concavite Aralo-Caspienne, Mediterranee et At-
lantique) tendent a masquer les grands traits de la constitution ge'ologique da
globe. (Get aper9u hasarde d'une ligne de commotion re'gulierement prolongee
n'exclut aucunement d'autres lignes selon lesquelles les mouvements peuvent se
pro pager e'galement.) " (Asie Centrale, t. ii. pp. 1 08. 120.) As the town of Khotan
and the country south of the Thian - schan were the most celebrated and most au-
cient seats of Buddhism, Buddhistic literature occupied itself early and much with
the causes of earthquakes. (See Foe-koue-ki, ou Kelation des Koyaumes Boud-
dhiques, trad, par Abel Remusat, p. 217.) The adherents of Sakhyamuni assign
eight such causes, among which the turning of a steel wheel hung round with
relics (s'arira) plays a principal part; a mechanical explanation, scarcely more
absurd than some geological and magnetical myths which were very late ia be-
coming obsolete among ourselves. According to a note of Klaproth'ti, mendicant
monks (Bhikchous) are imagined to have the power of making the earth tremble
and setting the subterranean wheel in motion. The travels of Fabian, the
author of the Foe-koue-ki belong to the beginning of the 5th century.

(219) p. 177. — Acosta, Viajes cientificos a los Andes ecuatoriales, 1849>
p. 56.

C250) p. 178.— Kosmos, Bd. i. S. 214—217 and 444 (English edition, pp. 194
— 197 and 426) , Humboldt, Eel. hist. t. iv. chap. 14, pp. 31 — 38. Ingenious
theoretical Considerations on Waves of Sound through the Earth, and Waves of
Sound through the Air, by Mallet, will be found in the Report of the Meeting of
the Brit. Assoc. in 1850, pp. 41—46, and in the Admiralty Manual, 1849,
pp. 201 and 217. In the tropics the animals which, according to my experience,
are earlier than human beings disquieted by the slightest earthquake movements,
are — poultry, swine, dogs, asses, and crocodiles (caymanes), which latter sud-
denly leave the bottom of the rivers.

.(MI) p. 179. — Julius Schmidt, in Noggerath, on the Earthquake of the 29th of
July 1846, S. 28—37. With the rate of velocity of the Lisbon earthquake as as-
signed in the text, the equatorial circumference of the earth would have been gone
round in about 45 hours. Michell (Phil. Trans., vol. li. pt. ii. p. 572) found,

Ivi NOTES.

for the earthquake of the 1st of November 1 755, only a rate of 50 English miles
a minute («. e. only 4170, instead of 7464, Paris feet in a second). There may
be here partly incorrectness in the older observations, partly diversity of propaga-
tion route. — A passage of Proclus in the Commentary to Plato's Cratylus,
throws a remarkable light on the connection of the sea-god Neptune with earth-
quakes alluded to in the text in the same page. " The middle one of the three
gods, Poseidon, is the cause of motion to all, even to those things which are least
movable. As the originator of motion he is called 'Evvocriyaws; and in casting
lots for the Croniau Empire, the middle lot with the moving sea fell to his
share." (Crenzer, Symbolik und Mythologie, 1842, Th. iii. S. 260.) As the
Atlantis of Solon, and, as I suppose, the kindred story of Lyctonia, are " geological
myths," so both these supposed fragmentary earthquake-divided lands were re-
garded as under the dominion of Neptune, and as such opposed to the " Saturnian
continent." According to Herodotus (lib. ii. c. 43 and 50), Neptune was a Lybian
divinity and unknown in Egypt. On these relations, on the disappearance of the
Lybian Lake Triton by the effects of an earthquake, and the belief of the great
rarity of earthquakes in the valley of the Nile, compare my Examen critique de
> Geographic, t. i. pp. 171 and 179.

(l52) p. 182.— The explosions of the Sangai, or Volcan de Macas, succeeded
each other on an average every 13-4 seconds ; Wisse, Exploration du Volcan de
Sangai, in the Comptes rendus de 1'Acad. des Sciences, t. xxxvi. 1853, p. 720.
As an example of earthquake phenomena limited to a very small space, I might
also have adduced the account given by Count Larderel of the " Lagoni " in
Tuscany. The vapours, which contain borax or borassic acid, announce their
presence and their being about to burst forth through fissures, by the agitation
of the suiTounding rocks. (Larderel, Sur les Etablissemens industriels de la
production d'acide boracique en Toscane, 1852, p. 15.)

(^ p. 182. — I am glad to be able to cite an important authority in support
of the view which I have attempted to unfold in the text : " Dans les Indes,
1'oscillation da sol, due a une eruption de volcans, est pour ainsi dire locale, tan-
dis qu'un tremblement de terre, qui, en apparence du moins, n'est lie' a aucune
e'ruption volcanique, se propage a des distances incroyables. Dans ce cas on a
remarque* que les secousses suivaient de preYeYence la direction des chaines de
montagnes, et se sont principalement ressenties dans les terrains alpins. La
frequence des mouvemens dans le sol des Andes, et le peu de coincidence que
Ton remarque entre ces mouvemens et les Eruptions volcaniques, doivent ne'ces-
sairement faire pre'sumer qu'ils sont, dans le plus grand nonibre de cas, occa-
sionne's par une cause independante des vokans" (Boussingault, Annales da
Chiruie et de Physique, t. Iviiu 1835, p. 83.)

NOTES. Ivii

(M4) p. 184. — The following was the order of succession of these great natural
events :-—

27 Sept. 1796. — Eruption of the volcano of the Island of Guadalupe
after many years of repose.

Nov. 1796. — The volcano on the elevated plateau of Pasto, between the
small rivers of Guaytara and Juanambu, became active, and began perma-
nently to emit smoke.

14 Dec. 1796. — Earthquake and destruction of the town of Cumarm.

4 Feb. 1797. — Earthquake and destruction of Riobamba. On the same
morning, at a distance of nearly 200 geographical miles from Riobamba,
the smoke from the volcano of Pasto suddenly disappeared, without any
earthquake shocks being felt in its vicinity.

30 Jan. 1811. — First appearance of the Island of Sabrina in the group
of the Azores, near the Island San Miguel. The elevation of the island
preceded the fiery eruption, as in the case of that of the Island of Santorin
and the volcano of Jorullo. After six days' eruption of scoriae, the island
rose to a height of 300 feet above the surface of the sea. It was the third
appearance and disappearance (by subsidence) of the island, after intervals
of ninety-one and ninety-two years ; always nearly at the same spot.

May 1811 to April 1812. — Above 200 earthquake shocks at the island
of St. Vincent.

December 1811. — A countless number of earthquake shocks in the
valleys of the Ohio, Mississippi, and Arkansas rivers, until 1813. North
of Cincinnati, between New Madrid, Little Prairie, and La Saline, the
earth trembled almost every hour during several months.

December 1811. — A single shock at Caracas.

26th March 1812. — Earthquake and destruction of the town of Caracas.
The agitation extended beyond Santa Marts, to the town of Honda, and the
elevated plateau of Bugota, 540 geographical miles from Caracas. The
movement lasted until the middle of the year 1813.

30th April 1812. — Eruption of the volcano of St. Vincent; and at the
same time, at two in the morning of the same day, a tremendous subterra-
nean noise, like the loudest cannonade, was heard, in equal intensity, on the
coasts of Caracas, in the Llanos of Calabozo, and the Rio Apure, without
being accompanied by any earthquake (see p. 177 in text). The subter-
ranean noise was also heard in the island of St. Vincent j but, which is
very remarkable, stronger at some distance on the sea.
(-") p. 186. — Humboldt, Voyage aux Re'gions Equinoxiales, t. ii. p. 376.

Iviii

NOTES.

(8S8) p. 186. — In order to compare, in the tropics, the temperature of springs
at the points where they issue directly from terrestrial strata, with the tempe-
rature of large rivers flowing in open channels, I collect here the following
numbers from my journals: —

Rio Apure, lat. 7f°, temp. 81°.
Orinoco, between 4° and 8° lat., temp. 81°'5— 85°'3.
Springs breaking forth from granite, in the forest near the cataract
of Maypures, temp. 82°.

Cassiquiare, the arm of the Upper Orinoco, which forms the connection
with the Amazons, temp, only 75°'8.

Rio Negro, above San Carlos, scarcely 1° 53' north of the equator, only
74a8.

Rio Atabapo, 79°'l.

Orinoco near the entrance of the Atabapo, lat. 3° 50', 82°.
Rio Grande de la Magdalena, lat. 5° 12' to 9° 56', temp. 79°'9.
Amazons, S. lat. 5° 31', opposite to the Pongo of Rentema (in the Pro-
vincia Jaen de Bracamoros), little more than 1200 feet above the sea, only
72°-5.

Thus it appears that the great mass of water in the Orinoco approaches the
mean atmospheric temperature of the region around. When great inundations,
overflowing the savannahs, take place, the yellow-brown waters, smelling of
sulphuretted hydrogen, become heated up to 92°-8; I found such a temperature
jn the Lagartero, filled with crocodiles, east of Guayaquil. The earth becomes
heated, as in shallow rivers, by the effect of the sun's rays. For the many
reasons (constantly clouded skies, abundant rains, evaporation from thick forests,
and absence of hot tracts of sand on the banks) of the lower temperature of the
"coffee-brown waters" of the Rio Negro and the "white waters" of the Cassi-
quiare, see my Relat. hist. t. ii. pp. 463 and 509. In the Rio Guancabamba or
Chamaya, which falls into the Amazons near the Pongo de Rentema, I have even
found the temperature only 67° 6; its waters coming down with enormous rapid-
ity, out of the high lake of Simicocha, from the Cordillera. During my fifty-
two days' voyage up the Magdalena, from Mahates to Honda, I recognised
clearly, by means of repeated observations, that a rise of the level of the river
could be foretold, some hours previously, from the lowering of the temperature.
The cooling of the water took place sooner as the source of the supply of the
cold mountain-waters from the Paramos was approached. Warmth and water
move, so to speak, in opposite directions, and with very unequal velocity. When
the waters rose suddenly at Badillas, the temperature had previously sunk from
80°-6 to 740<3. As passengers, sleeping at night, with their baggage, on a low

NOTES. llX

sandy island or on the shore, may be endangered by the rapid rising of the river,
some importance attaches to an indication which may afford a warning. [The
author then remarks that, throughout his work of the Kosmos, he always uses
the Centigrade scale where the contrary is not expressly stated. In this trans-
lation, both in the present and preceding volumes, the temperatures (where
nothing is expressly stated to the contrary) have always been converted from the
Centigrade into Fahrenheit's scale.]

(-57) p. 186. — Von Buch, Physicalische Beschreibung der Canarischen
Inseln, S. 8; Poggendorff's Annalen, Bd. xii. S. 403; Bibliotheque Britannique,
Sciences et Arts, t. xix. 1802, p. 263; Wahlenberg, De Veget. et Clim. in Hel-
vetia Septentrional! observatis, p. Ixxviii. and Ixxxiv.; the same, Flora Carpa-
thica, p. xciv.: and in Gilbert's Annalen, Bd. xli. S. 115; Humboldt in the
Mem. de la Soc. d'Arcueil, t. iii. 1817, p. 599.

(258) p. 187. — De Gasparin in the Bibliotheque Univ., Sciences et Arts,
t.xxxviii. 1828, pp. 54, 113, and 264; Me'in. de la Soc. Centrale d'Agriculture,
1826, p. 178; Schouw, Tableau du Climat et de la Vege'tation de 1'Italie, vol. i.
1839, pp. 133—195 ; Thurmann sur la Tempe'rature des Sources de la Chaine da
Jura, compare'e a celle des Sources de la Plaine Suisse, des Alpes, et des Vosges,
in the Annuaire Me'te'orologique de la France pour 1850,' pp. 258 — 268. De
Gasparin divides Europe, in respect to the frequency of summer and autumn
rains, into two contrasted regions. A rich collection of materials is contained in
Kamtz's Lehrbuch der Meteorologie, Bd. i. S. 448—506. According to Dove
(in Poggend. Ann., Bd. xxxv. S. 376), in Italy, " at places having a chain of •
mountains to the North, the maxima of the monthly quantities of rain fall in
March and November; and where the mountains are to the South, in April and
October." The totality of the relations in respect of rain in the temperate zone,
as far as they can be comprised in one general view, may be stated thus : " The
winter rainy season, at the borders of the tropics, divides, as we recede from
them, more and more into two maxima connected by slighter falls of rain ; in
Germany, these two have become reunited in a summer maximum, and there
remains no trace of a dry or rainless season." Compare the section " Geother-
mik" in the excellent "Lehrbuch der Geognosie" of Naumann, Bd. i. (1850)
S. 41—73.

(259) p. 187. — Compare present volume, p. 45 (English edition, p. 45).

C280) p. 189. — Compare Kosmos, Bd. i. S. 182 and 427, Anm. 9 (English
edition, p. 165 and Note 139); and present volume, p. 40 and Note 41 (English
edition, p. 41 and Note 41).

C261) p. 190.— Present volume, p. 37 (English edition, p. 37).

IX NOTES.

C282) p. 190. — Mina de Guadalupe, one of the Minas de Chota; present
volume, p. 41 and 42.

C283) p. 190. — Humboldt, Ansichten der Natur, Bd. ii. S. 323.

(2M) p. 190. — Mine on the Fleuss, in the Moll-Thai.-, see Hermann and
Adolph Schlagintweit, Untersuch iiber die physikalische Geographic der Alpen,
1850, S. 242^273.

(885) p. 192. — The same, in their work on Monte Eosa, 1853, Cap. vi.
S. 212—225.
. (266 94.— Humboldt, Kleinere Schriften, Bd. i. S. 139 and 147.

(267) p. 194. — The same, S. 140 and 203.

(268) P- 1^7. — I here depart from the opinion of a great friend, and a physi-
cist of great merit, in respect to the terrestrial distribution of heat. See, on the
cause of the hot springs of Leuck and Warmbrunn, Bischof, " Lehrbuch der che-
mischen und physikalischen Geologie," Bd. i. S. 127 — 133.

(2W) p. 197'. — See, on the subject of this passage discovered by Bureau de la
Malle, Kosmos, Bd. i. S. 231—232 and 448, Anm. 79 (English edition, p. 211,
and Note 209). " Est autem," said St. Patricius, " et supra firmamentum cseli,
et subter terram, ignis atque aqua; et quae supra terram est aqua, coacta in
unum, appellationem marium ; quse vero infra, abyssorum suscepit ; ex quibus
ad generis humani usus' in terram velut siphones quidam emittuntur et scatn-
riunt. Ex iisdem quoque et thermae existunt quarum qua? ab igne absunt lon-
gius, provida boni Dei erga nos mente,frigidiores; quae vero proprius admodum,
ferventes fluunt. In quibusdam etiam locis et tepidae aquae reperiuntur, prout
nwjore ab igne intervallo sunt disjunctae." The words are thus in the collection,
" Acta Primorum Martyrum, Opera et Studio Theodorici Ruinart," ed. 2, Amste-
laedami, 1713, fol. p. 555. According to another account (A. S. Mazochii in
Vetus Marmoreum Sanctse Neapolitans Ecclesiae Kalendarium Commentarius,
vol. ii. Neap. 1744, 4to, p. 385), Patricius spoke to the following effect : "Nam
quas longius ab igne subterraneo absunt, Dei optimi providentia, frigidiores
erumpunt. At quae propiores igni sunt, ab eo fervefactae, intolerabile calore
prasditas promuntur foras. Sunt et alicubi tepidae, quippe non parum sed lon-
giuscule ab eo igne remotae. Atqui ille infernus ignis impiarum est animarum
carnificina ; non secus ac subterraneus frigidissimus gurges, in glaciei glebas
concretus, qui Tartarus nuncupatur." The Arabic name hammam el-Enf sig-
nifies " nose baths," and is taken, as Temple has before remarked, from the shape
of a neighbouring promontory, and not from any supposed particularly favourable
effects of the waters in maladies affecting the nose. The Arabic name has been
variously given: hammam 1'Enf or Lif, Ernmamelif (Peyssonel), la Mamelif

NOTES. Ixi

(Desfontaines). Compare Gumprecht, die Mineralquellen auf dem Festlande von
Africa (1851), S. 140—144.

(27°) p. 198.— Humboldt, Essai Politique sur la Nouv. Espagne, 2 erne ed. t.
iii. (1827) p. 190.

0") p. 199. — Relat. hist, du Voyage aux Regions Equinoxiales, t. ii. p. 98.
Kosmos, Bd. i. S 230 (English edition, p. 208). The hot springs of Carlsbad
also owe their origin to granite (Von Buch in Poggend. Ann., Bd. xii. S. 416);
as do the hot springs of Momay in Thibet, visited by Joseph Hooker, which break
forth 16,000 feet above the sea, with a temperature of 124°'8, near Changokhang
(Himalayan Journals, vol. ii. p. 133).

C"72) p. 199. — Boussingault, Considerations sur les Eaux thermales des Cor-
dilleres, in the Annales de Cliimie et de Physique, t. Hi. 1833, pp. 188 — 190.

(273) p. 200.— Captain Newbold, " On the Temperature of the Wells and Rivers
in India and Egypt" (Phil. Trans. 1845, pt. i. p. 127).

(274) p. 201. — Sartorius von Waltershausen, Physisch-geographische Skizze
von Island, init besonderer Riicksicht auf volkanische Erscheinungen, 1847, S.
128 — 132. (Physico-Geographic Sketch of Iceland, with especial regard to
Volcanic Phenomena.) Bunsen and Descloiseaux, in the Comptes rendus des
Se'ances de 1'Acad. des Sciences, t. xxiii. 1846, p. 935; Bunsen, in the Anna-
len der Chemie und Pharmacie, Bd. Ixii. 1847, S. 27 — 45. Lottin and Robert
had already ascertained that the temperature of the column of water in the
Geysir diminishes from below upwards. Among the 40 silex-containing springs,
which are situated near the great Geysir and the Strokr, one bears the name of
" the Little Geysir." Its column of water rises only from 20 to 30 feet. Ac-
cording to an account of Esoma de Kerb's, there is a small Geysir, sending forth
a column of water about twelve feet high, in the high lauds of Thibet, near the
Alpine lake Mnpham.

(275) p. 201. — There are of fixed constituents, in 1000 parts in the springs of
Gastein Tromrnsdorf only 0'303; in the Lb'wig, at Pfeffers, 0'291; and in the
Longchamp, at Luxeuil, 0'236 : while, on the other hand, in every 1000 parts
of the common spring water at Berne, there have been found 0'478 ; in the
Carlsbad Sprudel, 5'459; and in the Wiesbaden water even 7'454. Studer,
a Physikal. Geogr. und Geologic," 2nd edition, 1847, Cap. i. S. 92.

(278) p. 202. — " Les eaux chaudes qui sourdent du granite de la cordillere du
littoral (de Venezuela) sont presque pures ; elles ne renferment qu'une petite
quantite' de silice en dissolution, et du gaz acide hydro-sulfurique mele' d'un peu
de gaz azote. Leur composition est identique avec celle qui resulterait de 1'ac-
tion de 1'eau sur le sulfure de silicium." (Annales de Chimie et de Phys.
t Hi. 1833, p. 189.) On the large quantity of azote in the hot spring of Orense

Ixii NOTES.

(154°-4), see Maria Rubio, Tratado de las Fuentes minerales de Espana, 1853,
p. 331.

(2") p. 202. — Sartorius von Waltershansen, Skizze von Island, S. 125.

(278) ?• 202. — The distinguished chemist Morechini, at Home, had given the
proportion of oxygen in the waters of the spring of Nocera (2238 feet above the
sea) as 0.40; Gay Lussac (26 Sept. 1805) found it more exactly only 0.299.
We had previously found 0.31 in the rain water. Compare, on the nitrogen
mixed with the "acid springs" of Nexis and Bourbon, 1'Archambault, the older
writings of Anglade and Lougchamp (1834); and on carbonic acid exhalations
in general, see Bischoffs excellent investigations in his " Chem. Geologic," Bd. i.
S. 243— 350.

(279) p. 202. — Bunsen in Poggend. Ann., Bd. Ixxxiii. S. 257; Bischoff,
Geologie, Bd. i. S. 271.

(280) p. 203. — Liebig and Bunsen, Examination of the Aix-la-Chapelle sul-
phurous Springs, in the Annalen der Chemie und Pharmacie, Bd. Ixxix. 1851,
S. 101. In chemical analyses of mineral waters containing sulphate of soda,
carbonic acid, soda, and sulphuretted hydrogen are often obtained, inasmuch
as there is in the same waters an excess of carbonic acid.

(281) p. 204. — One of these cascades is drawn in my " Vues des Cordilleres,"
PL xxx. On the analysis of the water of the Rio Vinagre, see Boussingault, in
the Annales de Chimie et de Phys., 2e sdrie, t. lii. 1833, p. 397; and Du-
mas, in the 3e se'rie of the same, t. xviii. 1846, p. 503 ; and, on the spring jn
the Paramo de Ruiz, Joaquin Acosta, " Viajes cientificos a los Andes ecuato-
riales," 1849, p. 89.

(2s2) p. 205. — The examples of altered temperature in the thermal waters of
.Mariura and las Trincheras lead us to ask whether the waters of the Styx, of
which the source, so difficult of access, is situated in the wild Arcadian Alps
near Nonacris, in the district of Pheneos, may have lost their noxious properties
by alterations which have taken place in subterranean fissures of communication?
—or whether they have only sometimes proved injurious to the traveller from their
icy coldness ? Possibly they may owe their evil renown, which is still handed down
.among the present inhabitants of Arcadia, and the myth which makes them take
their rise in Tartarus, solely to the awful wildness and desert character of tha
surrounding scenery. A young and accomplished philologist, Theodore Schwab,
succeeded a few years since, by great exertions, in reaching the rocky precipice
from which the source trickles down, just as is indicated by Homer, Hesiod, and
Herodotus. He drank of the exceedingly cold, and (judging by the taste) very
pure, water without finding that he could perceive any disadvantageous effect from
so doing (Schwab, Arkadien, seine Natur und Geschichte, 1852, S. 15—20). The

NOTES. Ixiii

ancients said that the extreme coldness of the waters of the Styx burst every kind
of vessel, except the hoof of an ass. Legends respecting the Styx are doubtless
of primeval antiquity, but the story of the poisonous qualities of its source appears
to have first become general in the time of Aristotle. According to the testimony
of Antigonus of Carystus (Hist. Mirab. § 174), it was contained, very circum,
stantially, in a lost book of Theophrastus. The calumnious fable of the poison-
ing of Alexander by Styx-water, which Aristotle had caused to come into tha
hands of Cassander, through Antipater, is refuted by Plutarch and Arrian ; it
•was circulated by Vitruvius, Justin, and Quintus Curtius, but without naming
Aristotle. (Stahr, Aristotelia, Tli. I. 1830, S. 137—140.) Pliny (xxx. 53)
says somewhat equivocally " Magna Aristotelis infamia excogitatum." Compare
Ernst Curtius, Peloponnesus, 1851, Bd. i. S. 194 — 196 and 212; St. Croix,
Examen crit. des anciens Historiens d'Alexandre, p. 496. A drawing of the Fall
of the Styx, taken from a distance, is given in Fiedler's Keise durch Griechert-
land, Th. i. S. 400.

C288) p. 205. — " Des gites me'talliferes tres-importants, les plus nombreux
peut-6tre, paraissent s'gtre forme's par voie de dissolution', et les filons concre-
tionnes n'etre autre chose que d'immenses canaux, plus ou moins obstrue's, par-
courus autrefois par des eaux thermales incrustantes. La formation d'un grand
nombre de mine'raux qu'on rencontre dans ces gites, ne suppose pas toujours des
conditions ou des agens tres-e'loignes des causes actuelles. Les deux e'le'mens
principaux des sources thermales les plus rdpandues, les sulfures et les carbonates
alcalins, m'ont suffi pour reproduire artificiellement, par des moyens de synthese
tres-simples, vingt-neuf especes minerales distinctes, presque toutes cristallise'es,
appartenant aux me'taux natifs (argent, cuivre et arsenic natifs); au quartz, au
fer oligiste, au fer, nickel, zinc et manganese carbonates; au sulfate de baryte, a
la pyrite, malachite, pyrite cuivreuse; au cuivre sulfure', a 1'argent rouge, arse-
nical et antimonial On se rapproche le plus possible des precedes

de la nature, si Ton arrive a reproduire les mine'raux dans leurs conditions d'as-
sociaticn possible, au moyen des agens chimiques naturels les plus re'pandus, et
en imitant les phe'nomenes que nous voyons encore se re'aliser dans les foyers ou
la cre'ation mine'rale a concentre* les restes de cette activite' qu'elle de'ployait
autrefois avec une toute autre dnergie." H. de Senarmont, sur la Formation
des Mine'raux par la Voie humide, in the Annales de Chimie et de Physique,
3e seVie, t. xxxii. 1851, p. 234. (Compare also Elie de Beaumont, sur les
Emanations volcaniques et me'talliferes, in the Bulletin de la Socie'te' Ge'ologique
de France, 2e se'rie, t. xv. p. 129.)

(**) p. 205. — " With the object of ascertaining the amount of difference of
the mean temperature of springs from the mean atmospheric temperature Dr.

IxiV NOTES.

Eduard Hallmann, at his former residence at Marienberg, near Boppard, on the
Rhine, observed during five years (from December 1, 1845, to November 30,
1850), the temperature of the air, the quantity of rain, and the temperature of
seven different springs; and founded, on these observations, a new and revised
examination of the question of the " Temperatur-Verhaltnisse der Quellen." He
excluded springs of thoroughly constant temperature (geological springs), in-
cluding only such as undergo a variation of temperature in the course of the
year. These " variable springs " fall under two natural groups : —

I) Purely meteorological springs, ». e., whose mean temperature is demon-
strably not heightened by terrestrial heat. In these, the amount of deviation
from the mean atmospheric temperature of the place is dependent on the distri-
bution of the fall of rain in the different seasons. Such springs are, on the
mean, colder than the air, when the quantity of rain falling in the four cold
months, from December to March inclusive, is more than a third of the whole
quantity falling in the entire year; and they are, on the mean, warmer than the
air, when the rain falling in the four warm months, July to October, exceeds
one third of the whole. The negative or positive departure from the atmospheric
mean is the greater, the greater the excess of rain in either the cold or the warm
season. Those springs, in which the deviation from the atmospheric mean is the
greatest and most accordant with the actual distribution of rain in the year, are
regarded as giving "undisfigured means;" while those, in which the deviation
from the atmospheric mean has been diminished by the disturbing influence of
the temperature of the air when no rain falls, are classed as giving " approximate
means " (both, however, coming under the general head of " purely meteorological
springs "). The too near approximation of the mean to the atmospheric mean
is a consequence either of the mode of obtaining the result — such as the
temperature being measured at the lower end of a channel or pipe coming from
the spring, — or of the spring itself having a superficial course, and of the
poverty of supply. Of the Marienberg springs, four belong to the group of
" purely meteorological springs," and of these four, one has an undisfigured
mean, and the other three, in different degrees, approximate means. In the
first observation-jear, the fall of rain was in excess in the cold third of the
year, and all the four springs were, in their means, colder than the air; in each
of the four following years, more than a third of the year's rain fell in the
wanner four months, and all the springs were, in their means, warmer than the
air, and the deviation was, in each year, proportionate to the amount of the
excess of warm rain.

Thus the view put forward by Von Buch in 1825, that the amount of devia-
tion of the mean temperature of springs from the mean atmospheric temperature

NOTES. Ixv

must depend on the distribution of the fall of rain within the annual period, has
been found perfectly correct by Hallmann, so far at least as his own station of
Marienberg, in the Rhenish greywacke, is concerned. It is only " purely meteo-
rological springs of undisfigured means " that are of value for scientific clima-
tology; such springs ought, therefore, to be sought for everywhere for this
object, in contradistinction from " purely meteorological springs of approxi-
mate means " on the one hand, and from " meteorolo-geological springs " on the
other.

2) Meteorolo-geological springs are such as have temperatures demonstrably
raised by the earth's heat. In these, whatever may be the distribution of the
fall of rain in the year, the mean is always higher than the atmospheric mean
(the variations which they show in the course of the year are imparted to them
by the ground through which they flow). The amount of their excess of mean
temperature over that of the air depends on the depth to which the meteoric
waters sink before reappearing as a spring, and has, therefore, no climatological
interest. The climatologist ought to know them, that he may not mistake them
for purely meteorological springs ; they may also, of course, present more or less
approximation to the atmospheric mean, by reason of any channel through which
their waters may be conducted to the point where their temperature is observed.
The springs were observed on fixed days, four or five times a month. The
height above the sea of the place of observation for atmospheric temperature, as
well as that of each of the different springs, was also noted.

Dr. Hallmann, after completing and reducing his Marienberg observations,
spent the winter of 1852 — 1853 in Italy; and found in the Apennines, besides
ordinary springs, some " abnormally cold springs." He gives this name to "springs
which can be shown to bring with them a lower degree of temperature from
heights. They are to be regarded as subterranean outlets, either from open lakes
situated high in the mountains; or from subterranean reservoirs, from which the
water descends in mass, and very rapidly, through cracks and fissures, and comes
out, at the foot of the hills, as a spring. The idea of an abnormally cold
spring is therefore, that it is too cold for the height at which it issues forth, or
rather that it issues forth at a part of the mountain too low down for its low
temperature." These views, given by Dr. Hallmann in the first volume of his
" Temperatur-Verhaltnisse der Quellen," are somewhat modified in the second
volume, S. 181 — 183, where he allows that all springs, even the most super-
ficial, contain some portion of warmth derived from the earth.

(285) p. 207. — Humboldt, Asie Centrale, t. ii. p. 58. Respecting the reasons
which render it more than probable that the Caucasus, which at f of its length,
between Kasbegk and Elburuz,in the mean parallel of 42° 50', runs E.S.E. —
VOL. IV e

Ixvi NOTES.

W.N.W., is a continuation of the volcanic fissure of the Asferah (Aktagh) and
Thian-schan, see the same, p. 54 — 61. Both these chains run, with little de-
viation, between the parallels of 40§° and 43°. The great depression of the Aral
and Caspian seas, the area of which depression is estimated by Struve, from exact
measurements, as exceeding that of the whole of France by almost 1680 square
(German) miles (Asie Centrale, t. ii. p. 309—312), is, I think, older than the
upheaval of the Altai and the Thian-schan. The fissure of elevation of the
last-named chain has not extended through the great depression. It only re-
appears west of the Caspian, as the chain of the Caucasus, with some variation
of direction, but with all trachytic and volcanic phenomena. This geological
connection is also recognised by Abieh, and is confirmed by him by important
observations. In a memoir on the connection of the Thian-schan with the
Caucasus, he says, expressly : — " The frequency and decided predominance, over
the whole district (between the Euxine and the Caspian), of a system of parallel
lines of dislocation and upheaval (nearly from east to west), carries the mean
direction of the axis of the great latitudinal Central Asiatic upheavals most
decidedly to the westward of the Kosyourt and Bolor systems, and to the
Caucasian isthmus. The mean direction of the Caucasus, S.E. — N.W., is, in
the central portion, E.S.E. — W.N.W., and sometimes even completely E. — W.,
like the Thian-schan. The lines of elevation, or upheaval, which connect
Ararat with the trachytic mountains Dzerlydagh and Kargabassar near Erze-
roum, and in the southern parallels of which Argasus, Sepandagh, and Sabalan
are arranged in line, are most decided indications of a mean volcanic axial
direction, L e. of a western prolongation of the Thian-shan through the Cau-
casus. Many other mountain directions of Central Asia are also met with again
in this remarkable space, and these, as everywhere else, are so linked together
as to form great mountain nodes, and maxima of elevation." Pliny (vi. 17)
says: — "Persas appellavere Caucasum montem Graucasim (var. Graucasum,
Groucasim, Grocasum), hoc est nive candidum;" wherein Bohlen thought he
recognised the Sanscrit words " kas " (to shine) and " gravan " (rock). (Com-
pare my Asie Centrale, t. i. p. 109.) If the name Graucasus has been cor-
rupted into Caucasus, it may, indeed, have happened, as Klausen remarks
in his Investigations on the Wanderings of lo (Rheinisches Museum fur Philo-
logie, Jahrg. iii. 1845, S. 298), that a name, in which each of the first syllables
recalled to the Greeks the idea of burning, may have denoted a burning moun-
tain, with which the story of the fire-kindler, TrupKaevs, easily connected itself
poetically." It is not to be denied that myths are sometimes occasioned by
names; but the origin of so great and important a myth as the Typhonian-
Caucasian can scarcely be derived from the accidental similarity of sound in a
misunderstood name of a mountain. There are better arguments, one of which

NOTES. Ixvii

is referred to by Klausen. From the conjunction of Typhon and Caucasus, and
from the express testimony of Pherecydes of Syros (in the fifty-eighth Olympiad)
it is clear that the eastern end of the world was supposed to be a volcanic moun-
tain. According to one of the Scholia to Apollonius (Scholia in Apoll. Rhod.
ed. Schgefferi, 1813, v. 1210, p. 524), Pherecydes said in the Theogony, "that
Typhon, being pursued, fled to Caucasus, and that there the mountain became
on fire, and that from thence he fled to Italy, where the island of Pithecusa was
thrown round him." Pithecusa is J^naria (now Ischia), in which island Mount
Epomeus (Epopon) emitted fire and lava, according to Julius Obsequens, ninety-
five years before our era; again, under Titus; again, under Diocletian; and
lastly, according to an exact account by Tolomeo Fiadoni of Lucca, at that time
prior of Santa Maria Novella, in 1302. Bb'ckh, who has so profound a know-
ledge of antiquity, said in a letter to me, " It is curious that Pherecydes should
make Typhon fly from Caucasus because it was on fire, he being himself the
author of the fire; but that his residence in the Caucasus was based on volcanic
eruptions believed to have taken place there, seems to me also undeniable."
Apollonius of Rhodes, where he speaks of the birth of the Colchian dragon
(Apollon. Rhod. Argon, lib. ii. v. 1212 — 1217, ed. Beck.), removes to the
Caucasus the " rock of Typhon," on which he was struck by the thunderbolt of
Zeus. Admitting the lava- streams and crater-lakes of the highland of Kely,
the eruptions of Ararat and Elburuz, and the streams of obsidian and lava from
the old craters of Riotandagh, to belong to pre-historic times; yet the numerous
cases of flames still breaking forth in the Caucasus, on mountains seven or eight
thousand feet high, as well as from fissures in wide plains, may well have
furnished ground enough for the whole Caucasian mountain- territory being
regarded as a " Typhonic seat of fire."

(286) p. 208. — Humboldt, Asie Centrale, t. ii. p. 511 and 513. I had
already remarked (t. ii. p. 201) that Edrisi does not mention the fire of Baku,
although 200 years before, in the tenth century, Massudi Cothbeddin had de-
scribed the country at length as a " Nefala-land," i e. rich in burning naphtha-
springs. (Compare Frahn, Ibn Fozlan, p. 245. and On the Etymology of the
Median word Naphtha, Asiat. Journal, vol. xiii. p. 124.)

(287) p. 209. — Compare Moritz von Engelhardfc and Fried. Parrot, Reise in
die Krym und den Kaukasus, 1815, Th. I. S. 71, with Gobel, Reise in die
Steppen des sudlichen Russland's, 1838, Th. I. S. 249—253; Th. II. S.
138—144.

(28S) p. 210. — Payen, De FAcide borique des Suffioni de la Toscane, in the
Annales de Chimie et de Physique, 3e se'rie, t. i. 1841, p. 247 — 255; Bischof,
Chem. und physik. Geologic, Bd. i. S. 669 — 691 ; Etablissements industries de
1'Acide boracique en Toscane, par le Comte de Larderel, p. 8.

Ixviii NOTES.

C280) p. 210. — Sir Roderick Ircpey Murchison, On the Vents of hot Vapour
in Tuscany, 1850, p. 7. (Compare also the earlier geological observations of
Hoffmann in Karsten's and Decken's Archiv fur Mineral. Bd. xiii. 1839, S. 19.)
Targioni Tozzetti states from old, but trustworthy, traditions, that some of
these springs or vents of boracic vapours were formerly seen to appear luminous
at night. In order to add to the geological interest of the considerations of Mur-
chison and Pareto, On the volcanic serpentine Formations in Italy, I would here
refer to the circumstance that the burning flame of the Chimaera in Asia Minor
(near the town of Deliktasch, the ancient Phaselis, in Lycia, on the west shore
of the Gulf of Adalia), which has burnt for thousands of years, also rises from a
hill on the declivity of Solimandagh, in which serpentine in situ and blocks of
limestone have been found. Bather more to the south, in the little island of
Grambusa, the limestone is seen vesting on dark-coloured serpentine. See the
valuable Memoir of Admiral Beaufort, Survey of the Coasts of Karamania, 1818,
p. 40 and 48, of which the statements have recently been entirely confirmed by
the rock-specimens brought home by a very gifted artist, Albrecht Berg. (Pierre
de Tchihatcheff, Asie Mineure, 1853, t. i. p. 407.)

C290) p. 210. — Bischof, work already cited, S. 682.

(891) p. 210. — Sartorius von Waltershausen, Physisch-geographische Skizze
von Island, 1847, S. 123; Bunsen, "Ueber die Processe der vulkanischen Ge-
steinsbildungen in Island," in Poggend. Anualen, Bd. Ixxxiii. S. 257.

C292) p. 210. — Waltershausen, work above cited, S. 118.

(293) p. 212. — Humboldt and Gay-Lussac, Mem. sur 1'Analyse de. 1'Air at-
mosphe'rique, in the Journal de Physique, par Lame'therie, t. Ix. an. 13, p. 151.
(Compare my Kleinere Schriften, Bd. i. S. 346.)

(891) p. 213. — " C'est avec e'motion que je viens de visiter un lieu que vous
avez fait connaitre il y a cinquante ans. L'aspect des petits volcans de Turbaco
est tel que vous 1'avez de'crit: c'est le meme luxe de la vegetation, le m§me
nombre ef la m§me forme des cones d'argile, la meme ejection de matiere
liquide et boueuse; rien n'est change' si ce n'est la nature du gaz qui se de'gage.
J'avais avec moi, d'apres les conseils de notre ami commun M. Boussingault,
tout ce qu'il fallait pour 1'analyse chimique des emanations gazeuses, merne pour
faire un melange frigorifique dans le but de condenser la vapeur d'eau, puis-
qu'on m'avait exprime' le doute, qu'avec cette vapeur on avait pn confondre 1'azote.
Mais cet appareil n'a e'te aucunement ne'cessaire. Des mon arrivee aux volcan-
citos, 1'odeur prononcee de bitume m'a mis sur la voie, et j'ai commence par
allumer le gaz sur 1'orifice meme de chaque petit cratere. On aperfoit meme
aujourd'hui, a la surface du liquide qui s'e'leve par intermittence, une mince
pellicule de pe'trole. Le gaz recueilli brule tout entier, sans residu d'azote (?)

NOTES. Ixix

et sans deposer du soufre (au contact de 1'atmosphere). Ainsi la nature du
phenomene a complement change depuis votre voyage, a moins d'admettre une
erreur d 'observation, justified par 1'etat moins avance de la cbimie expe'rimen-
tale a cette e'poque. Je ne doute plus maintenant que la grande eruption de
Galera Zamba, qui a eclaire le pays dans tin rayon de 100 kilometres, ne soit
mi phenomene de salses developpe sur une grande echelle, puisqu'il y existe.des
centaines de petits cfines, vomissant de 1'argile sale'e, sur une surface de plus de 400
lieues carrees. Je me propose d'examiner les produits gazeux dcs cSnes de Turbara,
qui sont les salses les plus eloignees de vos volcancitos de Turbaco. D'apres les
manifestations si puissantes qui ont fait disparaitre une partie de la pe'ninsule de
Galera Zamba, devenue une lie, et apres 1'apparition d'une nouvelle ile, soulevde
du fond de la mer voisine en 1 848 et disparue de nouveau, je suis porte a croire
que c'est pros de Galera Zamba, a 1'ouest du Delta du Eio Magdalena, que se
trouve le principal foyer du phenomene des salses -de la province de Carthagene."
(From a letter from Colonel Acosta at Turbaco to myself, 21 Dec. 1850.)
Compare also Mosquera, Memoria politica sobre la Nueva Granada, 1852, p.
73; and Lionel Gisborne's Isthmus of Darien, p. 4-8.

(295) p. 213. — Throughout the whole of my American expedition, I strictly
followed the advice given me by Vauquelin (under whom I had worked for some
time previously), to write down the details of every experiment the same day,
and to preserve the record. From my journals of the 17th and 18th of April,
1801, I subjoin the following extract: — " As, according to this, the gas, when
tried with phosphorous and nitrous gas, showed hardly O'Ol of oxygen, and
with lime-water less than 0'02 of carbonic acid, I ask myself, what are the re-
maining ninety-seven parts? I first surmised carburetted and sulphuretted
hydrogen ; but no sulphur is deposited on the margins of the little craters, in
contact with the atmosphere, nor was any smell of sulphuretted hydrogen per-
ceived. The problematical part might seem to have been pure nitrogen, since,
as above remarked, a lighted taper caused no ignition ; but I know from the
time when I made analyses in the Grubenwetter, that a light hydrogen gas, free
from any carbonic acid, which was merely near the roof of a gallery of the mine,
also did not ignite ; but, on the contrary, extinguished the miners' candles, which
burnt clear at points further in the interior, where there was a considerable
mixture of mephitic gas. The residue of the gas of the volcancitos may, there-
fore, be called nitrogen with a portion of hydrogen, of which we are not yet able
to state the quantity. May there be, under the volcancitos, the same carbonife-
rous schist, which I saw to the westward at Eio Sinu, or marl, and alum? l\Iay
atmospheric air penetrate through narrow cracks into cavities which have been
formed by water, and, in contact with dark-gray mud, become decomposed, as in

Ixx NOTES.

the works in the salt-clay of Wallein and Berchtholdsgaden ; where the chambers
become filled with gases which extinguish burning lights? Or do elastic gases,
in a state of tension, in flowing out, prevent the entrance of atmospheric air?"
I wrote these questions down, fifty-three years ago, at Turbaco. According to
the latest observations of M. Vauvert de Me'an (1854), the inflammability of
the gas exhaled has quite maintained itself. This traveller has brought back
samples of the water which fills the little crater-orifices of the volcancitos.
Boussingault found in a " litre" of it 6«r'59 of common salt; carbonate of soda,
0'31; sulphate of soda, 0'20; and also traces of borate of soda and iodine.
In the sediment deposited, Ehrenberg, by exact microscopic examination, re-
cognised no calcareous particles, and nothing belonging to scoriae; but grains of.
quartz, mixed with laminae of mica and many small crystalline prisms of black
augite, as it often presents itself in volcanic tufa; no trace of silicified sponges
or of poly gastric infusoria, nothing indicating the neighbourhood of the sea; but,
on the other hand, many remains of dicotyledonous plants, and of grasses, and
parts of lichens, reminding us of the constituents of the Moya of Pelileo.
Whereas Ch. Sainte-Claire Deville and Georg Bornemann found, in their fine
analysis of the Macalube di Terrapilata, O99 of carburetted hydrogen in the gas
emitted ; the gas which rises in the Agua Santa di Limosina near Catanea gave
them, as was formerly the case at Turbaco, 0'98 of nitrogen, without any trace
of oxygen. (Comptes-rendus de 1'Acad. des Sc. t. xliii. 1856, p. 361 and 366.)

(29G) p. 214. — Humboldt, Vues des Cordilleres et Monumens des Peuples
indigenes de I'Ame'rique, PI. XLI. p. 239. The fine drawing of the volcancitos
of Tnrbaco, from which the plate was engraved, was by my then young travel-
ling companion Louis de Rieux. On the ancient Tarruaco, in the earliest times
of the Spanish Conquista, see Hen-era, Dec. 1, p. 251.

(297) p. 215. — Lettre de M. Joaquin Acosta a M. Elie de Beaumont, in the
Comtes-rendus de 1'Acad. des Sc. t. xxix. 1849, p. 530 — 534.

C298) p. 216. — Humboldt, Asie Centrale, t. ii. p. 519 — 540, chiefly from
extracts taken from Chinese works by Klaproth and Stanislas Julien. The old
Chinese method of rope-boring, which, in the years 1830—1842, was on several
occasions employed with advantage in coal-mines in Belgium and Germany, had
been described (as Jobard discovered), in the seventeenth century, in the Relation
de 1'Ambassadeur Hollandais, Van Hoorn; but the most exact account of this
method of boring the Ho-tsing (fire-wells) was given by the French missionary
Imbert, who resided for many years at Kia-ting-fu. (Annales de 1' Association de
la Propagation de la Foi, 1829, p. 369 — 381.)

(2") p. 217. — According to Diard, Asie Centrale, t. ii. p. 515. Besides
the mud-volcanos at Damak and Surabaya, there are also, on other islands of

NOTES. Ixxi

the Indian Archipelago, the mud-volcanos of Pulu-Semao, Pulu-Kambing, and
Pulu-Roti. (See Junghuhn, Java, seine Gestalt und Pflanzendecke, 1852,
Abth. iii. S. 830.)

(30°) p. 218. — Junghuhn, same work, Abth. i. S. 201 ; Abth. iii. S. 854—
858. The weaker i( Dog-grottos " in Java are Gua-TJpas and Gua-Galan (the
first word in these names is the Sanscrit guha, grotto or cave). As it can
scarcely be doubted that the modern " Grotta . del Cane," near the Lago di
Agnano, is the same that was described by Pliny (ii. cap. 93), almost eighteen
centuries ago, " in agro Puteolano," as " Charonea scrobis mortiferum spiritum
exhalans," we can hardly refrain from joining in the wonder expressed by Scacchi
(Memorie geol. sulla Campania, 1849, p. 48), that, in a soil so loose, and so
often shaken by earthquakes, a phenomenon so minute as the conduction of this
small quantity of gas to its outlet, should have continued unaltered.

(301) p. 218. — Blume, Eumphia sive Commentationes botanies, t. i. (1835),
p. 47—59.

(302) p. 219. — Humboldt, Essai ge'ognostique sur le Gisement des Roches
dans les deux Hemispheres, 1823, p. 76; Boussingault, in the Annales de Chi-
rnie et de Physique, t. Iii. 1833, p. 11.

(303) p. 219. — On the height of Alausi (near Ticsan) on the Cerro Cuello
see the " Nivellement barometr." No. 206, in my Observ. astron. vol. i. p. 311.

(301) p. 220. — " L'existence d'une source de naphte, sortant au fond de la
mer d'un micaschiste grenatifere, et repandant, selon Texpression d'un historien
de la Conquista, Oviedo, une ' liqueur re'sineuse, aromatique et me'dicinale,' est
un fait extremement remarquable. Toutes celles que Ton connait jusqu'ici
appartiennent aux montagnes secondaires; et ce mode de gisement semblait
favoriser 1'ide'e que tous les bitumes mineraux (Hatchett dans les Transact, of
the Linncean Society, 1798, p. 129) e'taient dus h, la destruction des matieres
ve'ge'tales et animales ou a 1'embrasement des houilles. Le phenomene du Golfe
de Cariaco acquiert une nouvelle importance, si Ton se rappelle que le meme
teiTain dit primitif renferme des feux souterrains, qu'au bord des crateres en-
flamme's lodeur de petrole se fait sentir de temps en temps (p. e. dans 1 'eruption
du Ve'suve, 1 805, lorsque le Volcan Ian9ait des scories), et que la plupart des
sources tres-chaudes de 1'Ame'rique du Sud sortent du granite (las Trincheras
pres de Portocabello), du gneis et du schiste micace. Plus a Test du me'ridien
de Cumana, en descendant de la Sierra de Meapire, on rencontre d'abord le
terrain creux (tierra hueca) qui, pendant les grands trembiements de terre de
1766 a jetd de 1'asphalte enveloppe' dans du petrole visqueux; et puis au-delh, de
ce terrain une infinite' de sources chaudes hydrosulfureuses." (Humboldt, Relat.
hist, du Voyage aux Re'gions e'quin. t. i. p. 136, 344, 347, and 447.)

Ixxii NOTES.

(306) p. 223. — Kosmos, Bd. i. S. 244 (English edition, p. 223).

(306) p. 223. — Strabo, i. p. 58. Casaub. The epithet Sidirvpos shows that
mud-volcanoes are not here spoken of. Where Plato alludes to such in his
geognostical imaginations, in which he blends the mythical with the observed,
he says distinctly (in opposition to the phenomenon which Strabo has described)
vypov TnjAoD irorafMol. I have already treated, on another occasion (Bd. i.
S. 450—452, Anm. 95; English edition, Note 225), of the expressions ir-r]\6s
and /5uo£ as applied to volcanic outpourings. I will, therefore, only recall here
another passage in Strabo (vi. p. 269), in which the hardening lava, termed
Tr-r)\bs jUe'Aas, is most clearly characterised. In the description of Etna it is
said : " The glowing current (^a|) in hardening petrifies the surface of the
earth to a considerable depth, so that he who would uncover it must undertake
the labour of a quarryman. For as in the crater the rock is molten, and then
uplifted, so the fluid which streams from the summit is a black mass of mud
(7T7?A.(fc) which flows down the mountain, and after hardening becomes fit for
mill-stones, and retains the colour which it had before."

(307) p. 224. — Kosmos, Bd. i. S. 452, Anm. 98 (English edition, Note
228).

(30S) p. 224. — Leop. von Buch, Ueber basaltische Inseln und Erhebungs-
krater, in the Abhandl. der ion. Akademie der Wiss. zu Berlin, for 1818 and
1819, S. 51 ; and the same author's Physicalische Beschreibung der Canarischen
Inseln, 1825, S. 213, 262, 284, 313, 323, and 341. This work, which makes
an epoch in the well-based knowledge of volcanic phenomena, is the fruit of a
visit to Madeira and TenerifFe, from the beginning of April to the end of October,
1815. Naumann, however, notices very properly, in his Lehrbuch der Geogno-
sie, that in Von Buch's letters written from Auvergne in 1802 (Geognostische
Beob. auf Eeisen durch Deutschland und Italien, Bd. ii. S. 282), on the occasion of
the description of the Mont d'Or, the theory of elevation- craters, and their essen-
tial differences from volcanoes proper, were distinctly enounced. An instructive
companion-picture to the three craters of elevation in the Canaries (in the
islands of Palma, TenerifFe, and the Great Canary), is presented by the Azores.
The excellent maps of Captain Vidal, for the publication of which we are in-
debted to the British Admiralty, illustrate the wonderful geological conformation
of thos^ islands. In St. Michael we have the enormous Caldeira das sete
Cidades, formed almost under the eyes of Cabral in 1444, which is an elevation-
crater containing two lakes, the Lagoa Grande, and the Lagoa Azul, at a height
of 866 feet. The circumference of the Caldeira de Corvo, the dry portion of
whose floor is 1279 feet high, is almost as great. The elevation- craters of
Fayal and Terceira are situated at almost three times this height. To the same

NOTES. Ixxiii

kind of phenomena of eruption belong the numerous but transitory insular eleva-
tions which were seen, in 1691, in the sea round the island of S.Jorge, and, in
1757, round San Miguel, for a few days only. The periodical swelling of the
sea-bottom, three or four miles west of the Caldeira das sete Cidades, producing
a larger island, and which was of somewhat longer duration (Sabrina), has been
already noticed (Kosmos, Bd. i. S. 252; English edition, p. 230). On the
elevation-crater of Astruni in the Phlegraan Fields, and " the trachytic mass
pushed up in its centre as an unopened bell-shaped hill," see Leop. von Buch,
in Poggendorffs Annalen, Bd. xxxvii. S. 171 and 182. Rocca Monfina (mea-
sured and drawn in Abich's Geol. Beob. iiber die vulkan. Erscheinungen in Unter-
und Mittel-Italien, 1841, Bd. i. S. 113, Tafel II.) is a fine crater of elevation.

(309) p. 226. — Sartorius von Waltershausen, Physisch-geographische Skizze
von Island, 1847, S. 107.

(31°) p. 227. — It has been much debated what is the special locality in the
plain of Trcezene, or the peninsula of Methana, with which we should connect
the description of the Roman poet. My friend Ludwig Ross (the great investi-
gator of Grecian antiquity by the aid of many journeys) thinks that the im-
mediate neighbourhood of Troezene offers no locality which can be identified with
the inflated hill, and that Ovid must have transferred his graphic description to
that plain by a poetic license. Ross writes : " To the south of the peninsula
of Methana, and to the east of the Troezenian plain, lies the island of Ka-
lauria, known as the place where Demosthenes, pressed by the Macedonians,
took the poison in the temple of Poseidon. A narrow strait divides the limestone
of Kalauria from the coast, and gives the town and island their present name
(from Tt6pos, a passage). In the middle of the sound, connected with Kalauria
by a low, and possibly originally artificial, dike, there is a small conical island,
in shape like an egg cut through its length. It is throughout volcanic, and
consists of greyish yellow, and yello'wish and reddish, trachyte, with interspersed
lava and scoria;. It is almost entirely without vegetation. The present town
of Poros is built on it, on the site of the ancient Kalauria. Its formation appears
quite similar to that of the more recent volcanic islands in the bay of Thera
(Santorin). Ovid, in his spirited description, has probably followed a Greek
model, or an ancient legend." (Ludw. Ross in a letter to myself in November
1845.) Virlet, as a member of the French scientific expedition, expressed the
opinion that the volcanic elevation had been a later addition to the trachytic
mass of the peninsula of Methana, and was to be found at its north-western ex-
tremity, where the black burnt rock, called kammeni-petra, quite similar to the
kammeni at Santorin, betrays a later origin. Pausanias gives the tradition
among the inhabitants of Methana, that, on the north coast, before the still

1XX1V NOTES.

celebrated sulphureous thermal waters issued forth, fire had risen out of the
Earth. (See Curtius, Peloponnesos, Bd. i. S. 42 and 56.) On the " indescri-
bably sweet smell," which at Santorin (Sept. 1650) followed the sulphureous
stench, see Ross, Reisen auf den Griech. Inseln des iigaischen Meeres, Bd. i.
S. 196. On the smell of naphtha in the vapours from the lava of the Aleutian
island of Umnack, which appeared in 1796, see Kotzebue's Entdeckungs-Reise
Bd. ii. S. 106, and Leop. de Buch, Description physique des lies Canaries,
p. 458.

(3n) p. 228. — The highest summit of the Pyrenees, i. e. the Pic de Nethou
(the eastern and highest summit of the Maladetta or Malahita group), has been
twice measured trigonometrically, and its height above the sea is 3481 metres
(11,420 English feet) according to Reboul, and 3404 metres (11,168 English
feet) according to Coraboeuf. It is, therefore, 1705 feet lower than Mont
Pelvoux in the French Alps near Brian9on. In the Pyrenees, next after the
Pic de Nethou in height come the Pic Posets, or Erist, and, in the group of the
Marbore', Mont Perdu and the Cylindre.

(312) p. 228. — Me'moire pour servir a la Description ge'ologique de la France,
t. ii. p. 339. Compare on Valleys of Elevation and encircling Ridges in the
Silurian Formation, the excellent descriptions of Sir Roderick Murchison in his
Silurian System, Pt. I. p. 427—442.

(313) p. 228. — Bravais et Martins, Observ. faites au Sommet et au Grand
Plateau du Mont Blanc, in the Annuaire Me'te'orol. de la France pour 1850,
p. 131.

(•") pi 229. — Kosmos, Bd. iv. S. 221 (English edition, p. 172). I have
twice visited the volcanoes of the Eifel, at very different stages of the develop-
ment of modern geology: in the autumn of 1794, and in August 1845; the
first time in the district of the Laacher See, and the then still inhabited Abbey ;
and the second time, in the district round Bertrich, the Mosenberg, and the neigh-
bouring Maars; each time only for a few days. As, on the last excursion, I had
the happiness of accompanying my intimate friend Berghauptmann von Dechen,
I have been enabled, by a correspondence of many years, and by the communi-
cation of important manuscript memoirs, to avail myself freely of the observa-
tions of this clear-sighted geologist. I have often, as is my custom, distinguished,
by marks of quotation, passages taken verbally from his communications.

(315) p. 230. — H. von Dechen, Geogn. Uebersicht der Umgegend von Bad
Bertrich, 1847, S. 11—51.

(31fi) p. 230. — Stengel, in Noggerath, Das Gebirge von Rheinland und
Westphalen, Bd. i. S. 79, Tafel III. Compare also C. von Oeynhausen's ex-
cellent elucidations to his Geogn. Karte des Laacher Sees, 1847, S. 34, 39 und

NOTES. Ixxv

42 (embracing the Eifel and the Neuwied Basin). On the Maars, see Steininger,
Geognostische Beschreibung der Eifel, 1853, S. 113. His earliest, and merito-
rious, work, " Die erloschenen Vulkane in der Eifel und am Nieder-Rhein," be-
longs to 1820.

(317) p. 232. — Leucite (of the same sort as at Vesuvius, Rocca di Papa in
the Alban Hills, Viterbo, Rocca Monfina, — according to Pilla sometimes more
than three inches in diameter, — and in the Dolerite of Kaiserstuhl in the Breis-
gau) is also found " in situ as leucite-rock in the Eifel on the Burgberg near
Rieden. The tufa in the Eifel encloses great blocks of leucitophyr near Boll and
Weibern." I cannot resist the temptation of quoting from manuscript the fol-
lowing important remark of Mitscherlich's, contained in a chemico-geological
lecture given, a few weeks ago, in the Berlin Academy: — " It is only aqueous
vapours which can have caused the eruptions of the Eifel ; but these would have
divided the olivine and augite into the minutest drops, if they had found them
still fluid. In the main mass of the ejected substances, as for example at the
Dreiser Weiher, there are intermingled, in the most intimate manner, fragments
of the more ancient shattered rock, and which are frequently cemented together.
The great masses of olivine and the masses of augite are even, for the most part,
surrounded by a thick crust of this mixture ; one never finds in the olivine or in
the augite any fragment of the older rock ; both were, therefore, ready formed
before they came to the place where the shattering took place. The olivine and
augite had, therefore, already separated themselves from the fluid basaltic mass
before the latter came into contact with a collection of water or a spring, which
brought about an explosive ejection." Compare, respecting the bombs, an earlier
memoir by Leonard Homer, in the Transactions of the Geological Soc., 2nd
series, vol. iv. Pt. II. 1836, p. 467.

(318) p. 233. — Leop. von Buch, in Poggend. Annalen, Bd. xxxvii. S. 179.
According to Scacchi, these ejected substances belong to the first eruption of
Vesuvius of the year 79; Leonard's Neues Jahrbuch fiir Mineral., Jahrg. 1853,
S. 259.

(319) p. 236. — On the age of formation of the valley of the Rhine, see H.
von Dechen, Geogn. Beschr. des Siebengebirges, in the Verhandl. des naturhist-
Vereins der Preus. Rheinlande und Westphalens, 1852, S. 556—559. Ehren-
berg treats of the infusoria of the Eifel in the Monatsberichten der Akad. der
Wiss. zn Berlin, 1844, S. 337; 1845, S. 133 and 148; 1846,8.161—171.
The Trass of Brohl, filled with morsels of pumice containing infusoria, forms
hills, in some cases, more than 800 feet high.

(32°) p. 236. — Compare Roset, in the Me'moires de la Socie'te' Ge'ologique,
2e se'rie, t. i. p. 119. Also in the island of Java, that wonderful theatre of

Ixxvi NOTES.

varied volcanic activity, " craters without cones, as it were, flat volcanoes," are
found (Junghuhn, Java, seine Gestalt und Pflanzendecke, Lief. vii. S. 640)
between Gunung Salak and Perwakti, " as explosion-craters," analogous to
Maars. Without any encircling ridge, they are partly situated in entirely flat
parts among the mountains ; they have scattered around themselves angular
fragments of the rocky strata which have been exploded, and they now emit only
vapours and gases.

(321) p. 237. — Humboldt, Umrisse von Vulkanen der Cordilleren von Quito
und Mexico, a contribution to the Physiognomy of Nature, Plate IV. (Kleinere
Schriften, Bd. i. S. 133—205.)

C322) p. 237. — Umrisse von Vulkanen, Plate VI.

C52*) p. 237. — The same, Plate VIII. (Kleinere Schriften, Bd. i.' & ,463—
467. On the topographical situation of Popocatepetl (" Smoking Mountain " in
the Aztec language), in reference to that of the neighbouring "White Woman"
(Iztaccihuatl), and its geographical position relatively to the Lake of Tezcuco
on the west, and the Pyramid of Cholula on the east, see my Atlas ge'ogr. et
phys. de la Nouvelle Espagne, PL III.

(3M) p. 237. — Umrisse von Vulkanen, Plate IX.; the Star-Mount, in Aztec
language Citlaltepetl; Kleinere Schriften, Bd. i. S. 467— 470, and my Atlas
ge'ogr. et phys. de la Nouv. Espagne, PI. XVII.

(325) p. 237. — Umrisse von Vulk. PI. II.

(326) p. 237. — Humboldt, Vues des Cordilleres et Monuments des Peuples
indigenes de I'Ame'rique (fol.), PI. LXII.

(M7) p. 237. — Umrisse von Vulk. PI. I. and X. Kleinere Schriften, Bd. i.
S. 1—99.

C328) ^>. 238. — Umrisse von Vulk. PI. IV.

C820) p. 238. — The same, PI. III. and VII.

(33°) p. 238. — Long before the arrival of Bouguer and La Condamine (in
1736) in the high plain of Quito, long, therefore, before scientific measurements
of the heights of mountains, the natives knew that Chimborazo exceeded in
height all the other snow-clad mountains of the region. They had recognised
two lines of level which remain nearly constant throughout the year: one, the
the lower line of perpetual snow; and the other, the line to which a single casual
fall of snow descends. Inasmuch as in the equatorial region of Quito, as I have
elsewhere shown by measurements (Asie centrale, t. iii. p. 255), the snow-line,
on six of the highest of these mountain-giants, only varies about 200 feet ; and
as this variation, as well as the lesser ones produced by local circumstances, is
scarcely perceptible to the naked eye at so great a distance (the height of the
snow-line under the equator is equal to the height of Mont Blanc), the result is

NOTES. Ixxvii

an apparently perfect regularity and uninterrupted horizontally, which astonish
the observer accustomed to all the irregularity in the extent of the snowy cover-
ing which belongs to our variable temperate zone. This uniformity in the height
of the snow-line, and the knowledge of its maximum variation, in the region
round Quito, furnish an observer with vertical base-lines of 15,800 English feet
above the sea, and 6400 above the high plain ; of which, combined with very
exact measurements of angles of altitude, he can avail himself in the determina-
tion of distances, and the rapid solution of other topographical questions. The
second of the level lines above spoken of, i. e. the horizontal line which marks
the lowest limit to which the snow of a single casual fall descends, decides as
to the relative height of lower mountains, whose summits do not enter the region
of perpetual snow. Of a long chain of such summits, which had been erroneously
regarded as of equal elevation, many are seen to remain below the temporary fall
of snow while the rest are above it ; and thus the estimation as to their relative
height is corrected. In the mountain region of Quito, where the Sierras Nevadas
often approach each other, but without their perpetual snow-coverings being con-
tinuously connected, I have often heard, from the lips of uneducated country
people and herdsmen, such considerations and inferences from the lines of perpe-
tual snow and of casual snow showers. The grandeur of nature stimulates the
intellectual susceptibility of individuals among the coloured inhabitants, even
where they are in the lowest stage of civilisation.

(331) p. 239. — Abich, in the Bulletin de la Socie'te' de Ge'ographie, 4e se'rie,
t. i. (1851) p. 517; with a fine representation of the form of the ancient vol-
cano.

C02) p. 239. — Humboldt, Vues des Cord. p. 295, PI. LXL, and Atlas de
la Relat. hist, du Voyage, PI. XXVII.

(333) p. 240. — Kleinere Schriften, Bd. i. S. 61, 81, 83 and 88.

(33») p. 241. — Junghuhn, Eeise durch Java, 1845, S. 215, Tafel XX.

C335) p. 241. — See Adolf Erman's Eeise urn die Erde (a work of great geo-
logical importance, as well as valuable in so many other ways), Bd. iii. S. 271
'and 207.

(33G) p. 241. — Sartorius von Waltershausen, Physisch-geographische Skizze
von Island, 1847, S. 107; and his Geognostischer Atlas von Island, 1853, Taf.
XV. and XVI.

(337) p. 242. — Otto von Kotzebue, Entdeckungs-Eeise in die Siidsee und in die
Berings-Strasse, 1815—1818, Bd. iii. S. 68; Eeise- Atlas by Choris, 1820, Taf. V.;
Vicomte d'Archiac, Hist. desProgres de la Ge'ologie, 1847, t. i. p. 544 ; and Buzeta
Diccionario geogr. estad. historico de las Islas Filipinas, t. ii. (Madr. 1851), p. 436
and 470 — 471 : where, however, there is no mention of the double circle (of a second

Ixxviii NOTES.

crater in the crater-lake) which Delamare has described so circumstantially and
with so much scientific exactness in his letter to Arago (Nov. 18/42, Comptes ren-
dus de 1'Acad. des Sc. t. xvi. p. 756). The great eruption in Dec. 1754 (an earlier
and more violent one had taken place in September 1716) destroyed the old village
of Taal, situated on the south-western shores of the lake, and which has been sub-
sequently rebuilt further from the volcano. The small island in the lake, on
which the volcano rises, is called Isla del Volcan (Buzeta's work above referred
to). The absolute height of the volcano of Taal is about 890 feet. It is,
therefore, like Kosima, one of the very lowest. At the time of the American
expedition under Captain Wilkes (1842), it was in full activity. See United
States Explor. Exped. vol. v. p. 317.

(338) p. 242. — Humboldt, Examen crit. de 1'Hist. de la Ge'ogr. t. iii. p. 135;
Hannonis Periplus in Hudson's Geogr. Grseci Min. t. i. p. 45.

(339) p. 242. — Kosmos, Bd. i. S. 238 (English edition, p. 217).

(34°) p. 244. — For the situation of this volcano, which in smallness is only
exceeded by the volcanoes of Tanna and of the Mendana, see the fine Map of the
Japanese Empire by F. von Siebold, 1840.

(341) p. 244. — I do not here name, with the Peak of Teneriffe, among is-
land volcanoes, Mauna-roa, to whose conical form its name does not correspond ;
for in the Sandwich island language Mauna signifies mount, and rba signifies
long and very. Nor do I name Hawaii, the height of which was long debated,
and which was long described as an unopened trachytic dome. The celebrated
crater of Kiraueah, a, lake of molten lava, situated to the east of Mauna-roa,
near its foot, is 3969 feet high according to Wilkes. Compare the excellent de-
scription in Charles Wilkes's Exploring Expedition, vol. iv. p. 165 — 196.

(342) p. 245. — Letter from Fr. Hoffmann to Leop. von Buch, On the geolo-
gical Constitution of the Lipari Islands, in Poggend. Annalen, Bd. xxvi. 1832,
S. 59. Volcano, 1268 feet high according to the recent measurement of Ch.
Sainte Claire Deville, had strong eruptions of scoriaj and ashes in 1444, at the
end of the sixteenth century, in 1731, in 1739, and in 1771. Its fumaroles
contain ammonia, borate of selenium, sulphuret of arsenic, phosphorus, and,
according to Bornemann, traces of iodine. The three last-named substances are
here found for the first time among volcanic products. (Comptes Rendus de
1'Acad. des Sc. t. xliii. 1856, p. 683.)

(3J3) p. 245. — Squier, in the American Association (tenth annual meeting,
at New Haven, 1850).

(344) p. 245. — See Franz Junghuhn's highly instructive work, Java, seine
Gestalt und Pflanzendecke, 1852, Bd. i. S. 99. Ringgit is now almost extinct;
many thousand human beings perished in its terrible eruptions in 1586.

NOTES. Lxxix

(345) p. 245. — Thus the summit of Vesuvius is only 258 feet higher than
the Brocken.

(846) p. 245. — Humboldt, Vues des Cordilleres, PL XLIII.; and Atlas
ge'ogr. et phys. PL XXIX.

(347) p. 246. — Junghuhn, work above quoted, Bd. i. S. 68 and 98.

(348) p. 246. — Compare my Relat. hist. t. i. p. 93, particularly in respect
to the distance at which the summit of the volcano of the island Pico has some-
times been seen. The older measurement of Ferrer gave 7916 feet, therefore
304 feet more than the certainly more careful survey of Vidal in 1843.

(319) p. 246. — Erman, in his interesting geological description of the vol-
canoes of the peninsula of Kamtschatka, gives to Awatschinskaia or Gorelaia
Sopka 8910 feet; and to Strieloschnaia Sopka, also called Koriurkaia Sopka,
11,819 feet. (Reise, Bd. iii. S. 494 and 540.) Compare, respecting these two
volcanoes, of which the first is the most active, L. de Buch, Descr. phys. des lies
Canaries, p. 447 — 450. Erman's measurement of the volcano of Awatscha
agrees most nearly with the earliest measurement by Mongez, in 1787, in the
expedition of La Perouse (8757 feet), and with the more recent one of Beechey
(9056 feet). Hofmann in Kotzebue's, and Lenz in Lutke's, Voyages, found
only 8168 and 8212 feet. Compare Lutke, Voy. autour du Monde, t. iii. p. 67
— 84. Liitke's measurement of the Strieloschnaia Sopka gave 11,199 feet.

(sso) p> 246. — Compare Pentland's table of heights in Mary Somerville's
Phys. Geogr. vol. ii. p. 452; Sir Woodbine Parish, Buenos- Ayres and the Prov.
of the Rio de la Plata, 1852, p. 343; Poppig, Reise in Chile and Peru, Bd. i.
S. 411-434.

C351) p. 246. — May the summit of this remarkable volcano be diminishing
in height? A barometric measurement by Baldey, Vidal, and Mudge, in 1819,
still gave 9758 feet; whereas a very accurate and practised observer, who has
rendered important services to the geology of volcanoes, Sainte Claire Deville
(Voyage aux lies Antilles et a Tile de Fogo, p. 125), found, in 1842, only 9152
feet. Captain King, a short time before, had even found only 8810 feet.

(352) p. 247. — Erman, Reise, Bd. iii. S. 271, 275, and 297. The volcano
Schiwelutsch has, like Pichincha, the form, rare among active volcanoes, of a
long ridge (chrebet), on which rise detached domes and crests (grebni). Conical
and bell-shaped mountains are always designated, in the volcanic district of the
peninsula, by the term sopki.

(353) p. 247. — On the remarkable agreement of the trigonometric measure-
ment with the barometric one of Sir John Herschel, see Kosmos, Bd. i. S. 41,
Anm. 2 (English edition, Note 2).

(***) p. 247. — The barometric measurement of Sainte Claire Deville (Voy.

Ixxx NOTES.

aux Antilles, p. 102 — 118), in 1842, gave 12,158 feet, nearly accordant with
the result (12,181 feet) of the second trigonometric measurement of Borda in
1776, which I was enabled to publish from the Manuscrit du De'pot de la
Marine. (Humboldt, Vby. aux Re'gions e'quinox. t. i. p. 116 and 275 — 287.)
Borda's first trigonometric measurement, executed jointly with Pingre in 1771,
gave, instead of 12,181 feet, only 11,139 feet. The cause of the error was in
the erroneous notation of an angle (33' instead of 53'), as was related to me by
Borda himself, to whose great personal kindness I was indebted for so much
useful advice before my voyage to the Orinoco.

(**) p. 247. — I follow Pentland's statement of 12,367 feet; the more so
because Sir James Ross (Voy. of Discovery in the Antarctic Regions, vol. i.
p. 216) gave, as an approximate result, 12,400 feet as the height of this vol-
cano, of which the smoke and flames were visible even in the daytime.

(336) p. 247. — On Mount Argseus, which was first ascended and barome-
trically measured (at 12,705 feet) by Hamilton, see Pierre de Tchihatcheff,
Asie Mineure (1853), t. i. p. 441—449 and 571. William J. Hamilton, in his
excellent work (Researches in Asia Minor), obtained, as a mean result from a
barometric measurement and some angles, 13,000 feet; but if, as according to
Ainsworth, the height of Kaisarich is 1000 feet less than it was assumed by
Hamilton, it is only 12,000 feet. Compare Hamilton, in the Transact, of the
Geol. Soc. vol. v. Pt. III. 1840, p. 596. To the south-east of Argaeus (Erd-
schisch -Dagh), in the great plain of Eregli, south of the village of Karabunar
and of the mountain-group of Karadsha-Dagh, there rise several very small
cones of eruption. One of these, provided with a crater, has a form wonderfully
resembling that of a ship, running out in front into the shape of a prow. This
crater is in a salt lake, on the way from Karabunar to Eregli, fully four miles
from the former. The hill bears the same name. (Tchihatcheff, t. i. p. 455 ;
William J. Hamilton, Researches in Asia Minor, vol. ii. p. 217.)

(*") p. 247. — The height given is, properly speaking, that of the grass-
green mountain lake, the Laguna Verde, on the margin of which the Solfatara
examined by Boussingault is found. (Acosta, Viajes cientificos £ los Andes
ecuatoriales, 1849, p. 75.)

(3M) p. 247. — Boussingault reached the crater, and measured the height
barometrically ; it agrees very nearly with that which I stated by estimation,
twenty-three years earlier, on my journey from Popayan to Quito.

(**) p. 247. — Few volcanoes have had their height so over-estimated as
Mauna-roa. We have seen its supposed height gradually diminish from
18,400 feet (stated in Captain Cook's third voyage) to 16,482 feet in King's,
and 16,613 feet in Marchand's measurement, to 13,758 feet according to Wilkes,

NOTES. Ixxxi

and 13,528 according to Homer in Kotzebue's Voyage. The data, on which the
last-named result is based, were first published, by Von Buch, in the Descr.
phys. des lies Canaries, p. 379. Compare Wilkes, Explor. Exped. vol. iv.
p. 111—162. The eastern margin of the crater is only 13,428 feet. Mauna-
roa being stated to be without snow (lat. 19° 28'), the assumption of a greater
height would be in opposition to the result found by me in Mexico by many
measurements, giving, for the lower limit of perpetual snow, in the same latitude
14,470 feet. (Humboldt, Voy. aux Re'g. e'quinox. t. i. p. 97; Asie Centr. t. iii.
p. 269 and 359.)

(36°) p. 247. — This volcano rises to the west of the village of Cumbal, which
is itself 10,563 feet above the sea. (Acosta, p. 76.)

(361) p. 247. — I give the result of Erman's repeated measurements in Sept.
1829. The height of the margin crater must be subject to alterations from the
effects of the numerous eruptions; for measurements, equally carefully taken in
Aug. 1828, had given 16,029 feet. Compare Erman's Physikalische Beobach-
tungen auf einer Eeise urn die Erde, Bd. i. S. 400 and 419, with the " Histo-
rischen Bericht " of the Travels, Bd. iii. S. 358—360.

(362) p. 247. — Bouguer and La Condamine, in the Inscription at Quito,
give for Tungurahua before the great eruption of 1772, and before the earth-
quake of Riobamba (1797), which caused great mountain-falls, 16,773 feet: I
found trigonometrical ly, in 1802, only 16,490 for the summit of the volcano.

(363) p. 248. — The barometric measurement of the highest summit of the
Volcan de Purace, by Francisco Jose' Caldas (who, like my dear friend and
travelling companion Carlos Montufar, fell a sacrifice to his love for the in-
dependence and freedom of his native land), is given by Acosta (Viajes cientifi-
cos, p. 70) at 17,006 feet. For the little crater (Azufral del Boqueron), from
which sulphureous vapours issue with a violent noise, I found 14,414 feet.
(Humboldt, Eecueil d'Observ. astron. et d'Ope'rations trigon. vol. i. p. 304.)

(^^ p. 248. — Sangay is exceedingly remarkable for its uninterrupted acti-
vity, and for its situation : being somewhat to the east of the eastern Cordillera
of Quito, south of the Rio Pastaza, and 104 geographical miles from the nearest
part of the coast; therefore (like the volcanoes of the "Celestial Mountains" in
Asia) far from lending any support to the theories which say that the eastern
Cordilleras of Chili are free from volcanic eruptions by reason of their distance
from the sea. Darwin did not fail to remark, in some detail, on the subject of
this old and widely diffused coast-theory, in his Geological Observations on South
America, 1846, p. 185.

(36S) p. 248. — I measured the height of Popocatepetl, also called the Volcan
Grande of Mexico, in the plain of Tetimba, near the Indian village San Nicolas
VOL. IV. /

Jxxxii NOTES.

de los Ranches. It still appears to me uncertain which of the two volcanoes,
Popocatepetl or the Peak of Orizaba, is the highest. Compare Humboldt, Re-
cu'eil d'Observ. astron. .vol. ii. p. 543.

(s66) p. 248. — The Peak of Orizaba, covered with perpetual snow, of which,
previously to my journey the geographical position was given extremely erro-
neously upon all maps and charts notwithstanding its great nautical importance
to vessels coming to Vera Cruz, was first measured trigonometrically, in 1796,
by Ferrer from Encero. The result was 17,879 feet. I tried a similar opera-
tion in a little plain near Xalapa, and found only 17,374 feet; but the angles of
altitude were very small, and the base was difficult to level. Compare Hum-
boldt, Essai Politique sur la Nouv. Espagne, 2e eU t. i. 1825, p. 166; my
Atlas du Mexique (carte des fausses positions), PI. X. ; and Kleinere Schriften,
Bd. i. S. 468

f367) p. 248. — Humboldt, Essai sur la Ge'ogr. des Plantes, 1807, p. 153.
The height is uncertain, perhaps more than ^th too great.

(3G8) p. 248. — I measured, in 1802, the height of the truncated cone of the
volcano of Tolima, situated at the northern end of the Paramo de Quindiu, in
the Valle del Carvajal, near the little town of Ibague. The mountain is also
seen, at a great distance, from the high plain of Bogota. At this distance,
Caldas, in 1806, by a somewhat complicated combination, obtained a tolerably
approximate result, 18,430 feet; Semanario de la Nueva Granada, nueva edicion,
aumentada por J. Acosta, 1849, p. 349.

C69) p. 248. — The absolute height of the volcano of Arequipa has been so
variously given, that it is difficult to distinguish between mere estimations and
actual measurements. The distinguished botanistof Malaspina's Voyage of Circum-
navigation, Dr. Thadclaeus Hanke, of Prague, ascended it in 1796, and found on the
summit a cross, which had been ei'ected there twelve years before. By a trigonome-
trical operation, Hanke is said to have found, for the height of the volcano above the
sea, 20,335 English feet (19,080 Paris feet). This result, for the absolute height,
is much too great, and may perhaps have arisen from an erroneous assumption of
the absolute height of the town of Arequipa, in the neighbourhood of which the
operation was executed. If Hanke had then been provided with a barometer,
surely, after having been on the summit, he, as a botanist, unaccustomed to
trigonometrical operations, would not have resorted to them. The next person
who ascended the volcano, was Samuel Curzon, of the United States. (Boston
Philosophical Journal, 1823, Nov. p. 168.) In 1830, Pentland estimated it at
18,373 feet, and I have used this number (Annuaire du Bureau des Longitudes
pour 1'an 1830, p. 325) for my Carte hypsome'trique de laCordillere des Andes,
1831. It agrees pretty well (to almost ^) with the trigonometric measure-

NOTES. Ixxxiii

ment by a French naval officer, M. Dolley, kindly communicated to me in Paris
by Capitaine Alphonse de Moges, in 1826. Dolley found the summit of Are-
quipa 10,928 feet, and that of Charcani 11,857 feet above the high plain in
which the town of Arequipa is situated. Taking the height of the town at
7850 feet, from the barometric measurements of Pentland and Eivero (Pentland
gives 7852 feet in his table of heights, in the third edition of Mary Somerville's
Physical Geography, 3rd. ed. vol. ii. p. 454 ; and see Eivero, in the Memorial de
Ciencias naturales, t. ii. Lima, 1828, p. 65; Meyen, Eeise um die Erde, Th. II.
1 835, S. 5), we obtain, from Dolley's trigonometric operation, for the volcano of
Arequipa 18,876 feet, and for that of the volcano of Charcani 19,708 feet.
But the above-cited table of altitudes of Pentland gives for the volcano of Are-
quipa 20,320 feet (19,065 Paris feet), which is more than 2000 feet above
Pentland's estimate in 1830, and only too identical with Hanke's result in 1796,
which was 19,080 Paris feet.l On the other hand, in the Anales de la Uni-
versidad de Chile, 1852, p. 221, the same volcano is given so low as 18,373
feet ! A melancholy state of hypsometrical information !

(37°) p. 248. — Boussingault, accompanied by a highly informed companion,
Colonel Hall, almost reached the summit of Cotopaxi. They arrived, according
to barometric measurement, at a height of 18,862 feet. There remained only a
short distance to the margin of the crater, but the exceeding looseness of the snow
forbade further progress. Perhaps Bouguer's result may have been rather too
small, as his complicated trigonometric computation is dependent on the height
of Jhe town of Quito.

(371) p. 248. — Sahama, which Pentland (Annuaire du Bureau des Longi-
tudes pour 1830, p. 321) distinctly affirms to be a still active volcano, is situ-
ated, according to his new map of the valley of Titicaca (1848), east of Arica,
in the western Cordillera. It is 928 feet higher than Chimborazo; and the
proportion of its height to that of the lowest Japanese volcano Kosima, is as
30 to 1. I have abstained from placing in the fifth group the Chilian Acon-
gagua, which, according to its latest measurement, by Captain Kellett in the
Herald in 1845, is 23,004 feet high, because from the opposite opinions of
Miers (Voyage in Chili, vol. i. p. 283) and Charles Darwin (Journal of Ee-
searches into the Geology and Natural History of the different Countries visited
by the Beagle, 2nd ed. p. 291) it remains somewhat doubtful whether this
colossal mountain is a still active volcano. Mary Somerville, Pentland, and
Gillies (Naval Astron. Exped. vol. i. p. 127) deny its being so. Darwin says':
" I was surprised at hearing that Acongagua was in action the same night
(15th Jan. 1835), because this mountain most rarely shows any signs of
action."

1XXX17 NOTES.

C372) p. 249. — These breaking-through masses of porphyry show themselves
on a great scale near the Illimani in Cenipampa (15,946 feet) and Totoropampa
(13,705 feet). A quartz-porphyry, containing mica and enclosing garnets and
angular fragments of siliceous schist, forms the summit of the celebrated argen-
tiferous Cerro de Potosi. (Pentland, in MSS. of 1832.) The Illimani, to which
Pentland gave first a height of 7315, and afterwards of 6445 metres, has been
carefully measured, since 1847, by the engineer Pissis, who, on the occasion of
his great trigonometrical survey of the Llanura de Bolivia, found, by three tri-
angles between Calamarca and La Paz 6509 metres, 21,355 English feet, dif-
fering only 64 metres from Pentland's latest determination. See Investigaciones
sobre la Altitud de los Andes, in the Anales de Chile, 1852, p. 217 and 221.

(3ra) p. 250.—- Sartorius von Waltershausen, Geogr. Skizze von Island, S. 103
and 107.

(374) p. 251.— Strabo, lib. vi. p. 276, Casaub. ; Plin. Hist. Nat. iii. 9:
" Strongyle, quse a Lipara liquidiore flamma tantum differt; et cujus fumo qui-
nam flaturi sint venti, in triduo predicere incolas traduntur." Compare also
Urlich's Vindicias Plinianse, 1853, fasc. i. p. 39. The once so active volcano of
Lipara (in the north-east of the island) appears to me to have been either the
Monte Campo Bianco, or the Monte di Capo Castagno. (Compare Hoffmann, in
Poggend. Annalen, Bd. xxvi. S. 49 — 54.)

(375) p. 252. — Kosmos, Bd. i. S. 231 and 448, Anm. 77 (English edition,
vol. i. p. 210 and Note 207). Albert Berg, who had previously published a
picturesque work, Physiognomic der tropischen Vegetation von Sudamerika.
in 1853 went from Rhodes and the Bay of Myra (Andriace) to visit the Chi-
inaera in Lycia, near Deliktasch and Yanartasch. (The Turkish word tiisch,
signifies rock, as dagh and tagh do mountain; Deliktasch is " perforated rock,"
from the Turkish delik, a hole.) This traveller first saw the serpentine rock at
Adrasan. Beaufort had already seen, in the island of Garabusa (not Grambusa)
south of Cape Chelidonia, the dark serpentine rock resting on limestone (per-
haps imbedded in it). I subjoin an extract from a MS. communication from
Albert Berg : — " Near the ruins of an ancient temple of Vulcan rise the remains
OY a Christian church in the late Byzantine style; remains of a nave and of
two side chapels. In a fore-court to the east, the flame comes forth in an open-
ing like that of a fire-place, about two feet broad, and one foot high, in the ser-
pentine-rock. It rises about three or four feet high, and gives out a sweet smell,
perceptible at a distance of forty paces." (Is this a naphtha spring?) " Be-
sides this great flame, and beyond the hearth-like orifice, there also appear, on
side-clefts, very small, but constant, tongues of flame. The rock, where it is
touched by the flame, is much blackened ; and the soot deposited is collected for

NOTES. IXXXV

application to tlie eyelids, both to relieve pain, and particularly for colouring the
eyebrows. At three paces from the Chirnsera, the warmth from it is difficult to
bear. A stick of dry wood held in the opening near the flame, but without
touching it, takes fire. Where the ancient walls lean against the rock, gas
escapes from the interstices between the stones, which, probably because either
the temperature is lower, or the mixture is somewhat different, does not ignite
spontaneously, but burns when lighted. Eight feet below the large flame, in
the interior of the ruins, there is an opening, six feet deep, but only half as
wide, which probably was once vaulted over, for, in the damp part of the year, a
spring of water breaks out in it by the side of a cleft, over .which a little flame
plays." Berg shows, on a topographical plan, the geographical relations of the
alluvial strata, of the (tertiary ?) limestone, and of the serpentine-rock.

(37G) p. 253. — The oldest and most important notice on the volcano of
Masaya is in a manuscript, edited fourteen years ago by the meritorious histo-
rical collector Ternaux-Compans, of Oviedo's Historia de Nicaragua (cap. v. to
x.); see p. 115 — 197. The French translation forms a volume of the Voyages,
Relations, et Memoires originaux pour servir a 1'Histoire et a la Decouverte de
1'Ame'rique. Compare also Lopez de Gomara, Historia general de las Indias
(Zaragoza, 1553), fol. ex. b; and, among very recent writings, Squier, Nica-
ragua, its People, Scenery, and Monuments, 1853, vol. i. p. 211— 223, and
vol. ii. p. 17. The volcano of Masaya was then so celebrated that, in the Royal
Library at Madrid, there is a monograph upon it entitled: Entrada y Descubri-
miento del Volcan de Masaya que esta en la Prov. de Nicaragua, fecha por
Juan Sanchez del Portero. The author was one of those who let themselves
down into the -crater in the strange attempts of the Dominican Fray Bias de
Inesta. (Oviedo, Hist, de Nicaragua, p. 141.)

(3") p. 254. — In the French translation .of Ternaux-Compans (the Spanish
original has not been published), it is said, p. 123 and 132: — " On ne pent
cependant dire qu'il sorte pre'cisement une flamme du cratere, mais bien une
fume'e aussi ardente que du feu ; on ne la voit pas de loin pendant le jour, mais
bien de nuit. Le volcan e'claire autant que le fait la lune quelques jours avant
d'etre dans son plein." This remark, made so long ago on the kind of illumina-
tion of a crater, and of the strata of air over it, is not without bearing on the
doubts which have of late often been raised as to the disengagement of hydrogen
gas from the craters of volcanoes. If, in its ordinary state, here alluded to, El
Infierno de Masaya did not erupt scorias or ashes (Gomara adds: cosa que hazen
otros volcanes), yet it has sometimes had real eruptions of lava, of which pro-
bably the latest was in 1670. Since then the volcano has become completely
extinct, after a perpetual light from it having been observed for 140 years.

Ixxxvi NOTES.

Stephens, who ascended it in 1840, found no sensible trace of ignition. On the
Chorotega language, the meaning of the word Masaya, and the Maribios, see
Buschmann's ingenious ethnographic investigations Ueber die aztekischen Orts-
namen, S. 130, 140, and 171.

(878) p. 255. — "Les trois compagnons convinrent dedire qu'ils avaient trouve
de grandes richesses ; et Fray Bias, que j'ai connu comme un homme ambitieux,
rapporte dans sa relation le serment que lui et les associe's firent sur 1'e'vangile,
de persister h, jamais dans leur opinion que le volcan contient de 1'or m6le d'ar-
gent en fusion." Oviedo, Descr. de Nicaragua, cap. x. p. 186 and 196. The
chronicler de las Indias is very angry at Fray Bias having said that " Oviedo
had begged the emperor to give him El Infierno de Masaya in his coat of arms."
This would not, however, have been against the heraldic customs of the age;
for the brave Diego de Ordaz,who boasted of having reached the crater of Popo-
catepetl when Cortez first penetrated into the Valley of Mexico, received that
volcano as an addition to his arms, as did Oviedo the constellation of the
Southern Cross, and, earliest of all, Columbus (Examen crit. t. iv. p. 235 — 240)
a fragment of a map of the Antilles.

(379) p. 255. — Humboldt, Ansichten der Natur, Bd. ii. S. 276.

(39°) p. 256. — Squier, Nicaragua, its People and Monuments, vol. ii. p. 104.
(John Baily, Central America, 1850, p. 75.)

(391) p. 256. — Memorie geologiche sulla Campania, 1849, p. 61. I found
the height of the volcano of Jorullo 1680 feet above the plain in which it rose
and 4265 feet above the sea.

£»2) p 257. — La Condamine, Journal du Voyage a 1'Equateur, p. 163; the
same, in the Mesure de trois Degre's de la Me'ridienne de THdmisphere austral,
p. 56.

C383) P- 257. — At the country*house of the Marques de Selvalegre, the father
of my unfortunate companion and friend Don Carlos Montufar, one was often
inclined to attribute the bramidos — which resembled the firing of a distant
battery of heavy guns, and were heard with exceedingly unequal intensity, the
direction of the wind, and the state of the atmosphere, temperature, &c., remain-
ing unchanged, — not to Sangay, but to Guacamayo, a mountain forty miles
nearer, at the foot of which a path leads from Quito by the Hacienda de Anti-
sana to the plains of Archidona and the Rio Napo. (See my special map of the
Province of Quires, No. 23 of my Atlas ge'ogr. et phys. de 1'Amerique, 1814—
1834.) Don Jorge Juan, who heard the Sangay thunder much nearer than I
did, says decidedly that the bramidos, which he terms the ronquidos del volcan
(Eelacion del Viage a la America Meridional, Pt. I. t. ii. p. 569), belong to
Sangay, or Volcan de Macas, whose voice, if I may use the expression, is very

NOTES. Ixxxvii

distinctly characterised. To the Spanish astronomers this voice appeared par-
ticularly hoarse, and they, therefore, called it a snore (un ronquido), rather than
a roar (bramido). The mysterious noise of the volcano Pichincha, which I
heard several times at night at Quito without its being followed by any earth-
quake, has something clear and ringing, as if chains were made to rattle, or as
if masses of glass fell upon each other. Wisse describes the noise of Sangay,
when at the mountain itself, as sometimes like rolling thunder, and sometimes
detached and sharp, as if one was in the midst of near platoon firing. From the
top of Sangay to Payta and San Buenaventura in Choco, where the bramidos of
Sangay are heard, the distances, in a south-west direction, are respectively 252
and 348 geographical miles. (Compare Carte de la Prov. de Choco, and Carte
hypsome'trique des Cordilleres, No. 23 and 3 of my Atlas ge'ogr. et phys.)
Thus in these regions of wild grandeur, including Tungurahua and Cotopaxi,
whose loud noise was heard by me from the Pacific in February 1803 (Kleinere
Schriften, Bd. i. S. 384), the voices of four volcanoes may be heard from places
but little distant from each other. The ancients speak of the difference of the
noise heard in the ^Eolian islands from the same fiery orifice at different times.
(Strabo, lib. vi. p. 276.) In the great eruption (January 23, 1835) of the
volcano of Conseguina, situated on the shore of the Pacific, at the entrance of
the Gulf of Fonseca, in Central America, the subterranean propagation of sound
was such that the noise was heard distinctly on the high plain of Bogota: a dis-
tance like that of Etna from Hamburg. (Acosta, in the Viajes cientificos de M.
Boussingault a los Andes, 1849, p. 56.)

884) p. 258. — Kosmos, Bd. iv. S. 230 (English edition, p. 182).
(s85) p. 260. — Compare Strabo, lib. v. p. 248, Casaub., e%6t Koi\ias Tivds,
and lib. vi. p. 276. The geographer of A masia expresses himself, with much
geological sagacity, on two different kinds of origin in islands (vi. p. 258): —
" Some islands (naming them) are fragments of the main land ; others pertain
more to the sea; for these latter islands of the high seas (lying far out at sea)
were probably raised up from the deep ; whereas those lying off promontories,
and separated only by a narrow strait, may more reasonably be supposed to have
been torn off from the mainland." (This is from Groskurd's German version.)
The small group of the Pithecusaj consisted of Ischia, probably originally called
JSnaria, and Procida (Prochyta). Why this group should have been supposed
to have had apes in it in ancient times, and should have been named from this
supposition by the Greeks and the Italian Tyrrhenians — Etruscans therefore —
(Strabo, lib. xiii. p. 626 : apes in Tyrrhenian were &pi/j.oi'), remains very obscure,
and may perhaps have been connected with the fable of the former inhabitants
having been changed into apes by Jupiter. The word &pi/j.oi, for apes, recalls

Ixxxviii NOTES.

the Arimse of Homer, II. ii. 783, and of Hesiod, Theog. v. 301. The words fii>
api/j-ois, of Homer, are, in some codices, contracted into one, and it is thus we
find the word in Koman writers (Virg. JEn. ix. 716; Ovid. Metam. xiv. 88).
Plinius (Hist. nat. iii. 5) even says positively: "JSnaria, Homero Inarime dicta,

Gra3cis Pithecusa " The Homeric land of the Arimes, where Typhon

lay, was looked for, in antiquity itself, in Cilicia, Mysia, Lydia, in the volcanic
Pithecusse, in the crater Puteolanus, and in the Phrygian Burnt Land (under
which Typhon once lay), and even in the Katakekaumene. That apes should
have lived in Ischia, so far from the African coast, within historic times, seems
the more improbable, because, as I have elsewhere remarked, it is not even
proved that apes lived in ancient times on the rock of Gibraltar; for Edrisi (in
the twelfth century), and other Arabian geographers who have described the
Straits of Hercules so circumstantially, do not mention them. Also Pliny denies
the apes in -3£naria, but deduces the name of the PithecusaB, in the most impro-
bable manner, from iridos, dolium (a figlinis doliorum). Bockh says: — "It
seems to me that the principal thing in this inquiry is that Inarima is a name
of the Pithecusse, which has arisen out of learned interpretations and ficuon, as
Corcyra would in this way become Scheria; and that ^Eneas was first connected
with the Pithecusse (jEnese insulse) by the Eomans, who found their ancestor
everywhere in these regions. Naevius also seems to bear witness to this connec-
tion with tineas in the first book of the Punic War."

(*6) p. 260. — Pind. Pyth. i. 31. Compare Strabo, v. p. 245 and 248, xiii.
p. 627. We have already remarked (p. 207, Note 285) that Typhon fled from
the Caucasus to Lower Italy : as if the myth indicated that the volcanic erup-
tions in the latter country were less old than in the Caucasian isthmus. The
consideration of popular myths cannot be separated from the geography and the
history of volcanoes. They often throw light reciprocally upon each other. What
would be regarded as the most powerful of the moving forces on the surface of
the Earth (Aristot. Meteorol. ii. 8, 3), the wind, the enclosed pneuma, would
be recognised as the general cause of volcanic action (fire-emitting mountains
and earthquakes). Aristotle's consideration of nature was founded on the reci-
procal action of the external and internal subterranean air, on a theory of exha-
lation, on differences of heat and cold, of moist and dry. (Aristot. Meteor, ii.
8, 1, 25, 31, and ii. 9, 2.) The greater the mass of the winds enclosed in
subterranean and submarine hollow passages, and the more they are hindered
from irioving rapidly and far according to their natural inherent property, the
more violent will be the eruptions. " Vis fera ventorum, csecis inclusa cavernis."
(Ovid. Metam. xv. 299.) Between the " pneuma " and the " fire " there is a
peculiar relation. (Tb irvp OTCLV /*erct Trveu/iaros 77, yivtrcu

XOTES. Ixxxix

, Aristot. Meteor, ii. 8, 3; KCU yao T& irvp olov TTVCVUCITOS ns Averts,
Theophrast., De Igne, § 30, p. 715.) Also from the clouds, the pneuma sud-
denly set free sends forth the kindling and far shining thunderbolt (TTPTJITT^P).
" In the Land of Fire, the Katakekaumene of the Lydians," says Strabo (lib.
xiii. p. 628), "there are still shown three hollows, fully forty stadia from each
other, which are called the Bellows; there lie over them rough hills, probably
piled up by the glowing masses which have been blown out." He had previously
said (lib. i. p. 57) that "between the Cyclades (Thera and Therasia) flames of
fire had for four days burst forth from the sea, so that the whole sea boiled and
burnt; and gradually an island formed of glowing masses was heaved up as by
a lever." All these phenomena, so well described, are attributed to the com-
pressed winds, acting as elastic vapours. Ancient physics concerned themselves
but little with the several varieties of substances; they were dynamic, and
regarded the measure of motive force. The view. of the increasing warmth of
the Earth with increasing depth as the cause of volcanoes and earthquakes, is
first found expressed, in the third century, by a Christian bishop in Africa, under
Diocletian. (Kosmos, Bd. iv. S. 244; English edition, p. 197.) The Pyri-
phlegethon of Plato, as a stream of fire circulating in the interior of the earth,
feeds all lava-giving volcanoes, as noticed above in the text, p. 261. ,In the
earliest presentiments of men, and within a small circle of ideas, we find the
germs of that which we now think we can explain under the form of other
symbols.

(W7) p. 262. — Mount Edgecombe, or Mount St. Lazarus, on the small
island (Croze's island) which lies west of the larger island of Sitka or Baranow
in Norfolk Sound, and which had already been seen by Cook, is a hill of only
2770 feet high, composed partly of basalt rich in olivine, and partly of felspathic
trachyte. Its last great eruption, which brought much pumice to the surface,
was in 1796. (Lutke, Voyage autour du Monde, 1836, t. iii. p. 15.) Eight
years afterwards, Captain Lisiansky went to the summit, where there is a
crater lake. He found at that time no indications of activity at any part of the
mountain.

(**) p. 264. — As early as under the Spanish dominion, in 1781, the
Spanish engineer, Don Jose' Galisteo, had found, for the surface of the Laguna
of Nicaragua, a height only six feet above that found by Baily in his different
levellings in 1838. (Humboldt, Rel. hist. t. iii. p. 321.)

(389) p. 265. — Compare Sir Edward Belcher, Voyage round the World,
vol. i. p. 185. I found myself in the Papagayo storm, according to my chrono-
metric longitude 19° 11' West of the meridian of Guayaquil, or in 99° 07' W.
of Greenwich, 880 geographical miles west of the shore of Costa Rica.

XC NOTES.

(88°) p. 265. — My earliest writing on the subject of a range of seventeen
Guatemala and Nicaragua volcanoes is in the geographical Journal of Berghaus
(Hertha, Bd. vi. 1826, S. 131 — 161). At that time I could avail myself, in'
addition to the old chronicles of Fuentes (lib. ix. cap. 9), only of the important
Compendio de la Historia de la Ciudad de Guatemala, by Domingo Iguarros, and
of three maps, by Galisteo (from a survey made by order of Matias de Galvez,
Viceroy of Mexico, 1781), by Jose Rossi y Rubi (Alcalde Mayor de Guatemala,
1800), and by Joaquin Ysasi and Antonio de la Cerda (Alcalde de Granada),
which I possess, principally in manuscript. Leopold von Buch, in the French
version of his work on the Canary islands (Descr. phys. des lies Canaries, 1836,
p. 500 — 514), enlarged my sketch in a masterly manner; but the uncertainty
which then prevailed respecting geographical names, and which led, in some
cases, to one name being confounded with another, gave rise to many doubts ;
these have now, for the most part, been set at rest by the fine map of Baily and
Saunders, by Molina's Bosquejo de la Republica de Costa Rica, and by the great
and very meritorious work of Squier (Nicaragua, its People and Monuments,
with Tables of the comparative Heights of the Mountains in Central America,
1852; see vol. i. p. 418, and vol. ii. p. l62). The important work promised to
us by Dr. Oersted, under the title " Schilderung der Naturverhaltnisse von
Nicaragua und Costa Rica," in addition to the botanical and zoological investi-
gations which were the principal objects of the author's journeys, will also throw
light on the geology of Central America ; which he traversed, in a variety of
directions, from 1846 to 1848, and from whence he brought back with him to
Copenhagen a collection of rock-specimens. I owe to his kind communications
some interesting rectifications of my fragmentary work. From a careful com-
parison of the materials which I now possess, including also some very valuable
ones from the Prussian Consul-General in Central America, Herr Hesse, the
following table of the volcanoes of Central America has been drawn up, proceed-
ing from South to North : —

On the central high plain of Cartago (4646 feet), in the republic of Costa
Rica (N. lat. 10° 9'), rise the three volcanoes of Turrialva, Irasu, and Reventado
of which the two first are still active.

Volcan de Turrialva* (about 11,000 feet) is, according to Oersted, only
divided from Irasu by a deep narrow cleft. Its summit, from which columns of
smoke rise, has not been ascended.

Volcan de Irasu*, also called the volcano of Cartago (11,096 feet), to the
north-east of the volcano of Reventado, is the principal seat of volcanic activity
in Costa Rica; yet it is remarkably accessible; on the southern side there is a
succession of terraces, the summit, from whence both oceans are visible, may

NOTES. XC1

be reached almost entirely on horseback. The cone of cinders and rapilli, about
1000 feet high, rises out of an encircling ridge (a crater of elevation). On the
flatter north-eastern portion of the summit lies the proper crater, between 7000
and 8000 feet in circumference, and which never sent forth streams of lava. Its
eruptions of scoriae have often been accompanied, in 1723, 1726, 1821, and
1847, by earthquakes which have destroyed towns, and have extended from
Nicaragua or Rivas to Panama. (Oersted.) Dr. Carl Hoffmann, in a recent
ascent of Irasu, in May 1855, examined the summit crater and the orifices of
eruption more closely. The height of the volcano is, according to a trigono-
metric measurement by Galindo, 12,000 Spanish (11,000 English) feet. (Bon-
ulandia, 1856, No. 3.)

El Reventado (9486 feet), with a deep crater, whose southern margin has
fallen in, and which was formerly filled with water.

Barba (about 8500 feet), north of San Jose', the capital of Costa Rica, with a
crater which encloses several little lakes.

Between the volcanoes of Barba and Orosi there is a line of volcanoes, which,
running east and west, crosses the principal chain, of which the direction in Ni-
caragua and Costa Rica is S.E. — N.W. On this cross-fissure rise, beginning
from the east, Miravalles and Tenorio (each about 4700 feet high) ; in the mid-
dle, south-east of Orosi, Rincon, also called Rincon de la Vieja* (Squier, vol. ii. p.
102), which every spring, at the beginning of the rainy season, has small erup-
tions of ashes ; and most to the west, near the little town of Alajuela, the vol-
cano of Votos *, rich in sulphur (7513 feet). Dr. Oersted compares this phaeuo-
menon of the direction of volcanic activity over a cross-fissure, with the east and
•west direction, which I found in the Mexican line of volcanoes from sea to sea.

Orosi*, still active, in the most southern part of the State of Nicaragua (5220
feet) ; probably the Volcan del Papagayo on the chart in the Deposito Hidro-
grafico.

The two volcanoes of Mandeira and Ometepec* (4160 and 5220 feet), on a
small island in the Laguna of Nicaragua, called by the Aztec inhabitants of the
district, " ome teptl" or "two mountains." (Compare Buschmann, Aztekische
Ortsnamen, S. 178 and 171.) The island-volcano Ometepec, wrongly called by
Ignarros, Ometep (Hist, de Guatem. t. 1, p. 51), is still active. It is drawn in
Squier, vol. ii. p. 235.

The extinct crater of the island Zapatera, but little raised above the level of
of the sea. The period of its former eruptions is wholly unknown.

The volcano of Momobacho, on the western shore of the Laguna de Nicaragua,
a little to the south of the town of Granada. As this town (which is also called
Mombacho, Oviedo. Nicaragua, ed. Ternaux, p. 245) is situated between this vol-

NOTES.

cano and Massaya; pilots designate sometimes one and sometimes the other of
these mountains by the indefinite name of the Volcano of Granada.

Massaya (Masaya) has been described more fully in the text, p. 253 — 255.
It was formerly as active as Stromboli, but has been extinct since the great erup-
tion of lava in 1670. According to the interesting account by Dr. Scherzer
(Sitzungsberichte der philos. hist. Classe der Akad. der Wiss. zu Wien, Bd. xx.
S. 58), in April 1853, clouds of steam were again emitted from a newly opened
crater. The volcano of Massaya is situated between the two lakes of Nicaragua
and Managua, to the west of the town of Granada. Massaya is not synonymous
•with Nindiri; but Massaya and Nindiri * form a " twin volcano," with two summits
and two different craters, both of which have given out lava. The stream of lava
from Nindiri in 1775 reached the lake of Managua. The equal height of the
two volcanoes so near to each other is stated to be only 2450 feet.

Volcan de Momotombo* (7030 feet), active ; subterranean thunder often
heard, without smoke being seen ; in N. lat. 12° 28', at the north end of the
•Laguna de Managua, over against the small island of Momotombito, rich in
sculptured remains. (See the drawing of Momotombo in Squier, vol. i. p. 233
and 302 — 312.) The Laguna de Managua is between 27 and 28 feet higher
than the Laguna de Nicaragua, which is more than twice as large, and has no
island-volcano.

From hence to the Gulf of Fonseca or Conchagua, at a distance of twenty
miles from the coast line of the Pacific, there extends a line running S.E. — N.W.
of six volcanoes, near to each other, and bearing the common name of Los Mari-
bios. (Squier, vol. i. p. 419, vol. ii. p. 123.)

El Nuevo*, wrongly called Volcan de las Pilas, because the eruption of the
12th of April 1850, took place at the foot of that mountain; a considerable
eruption of lava almost in the plain itself ! (Squier, vol. ii. p. 105—110.)

Volcan de Telica*, visited in the 16th century (about 1529) by Oviedo,
during its activity ; situated to the east of Chinendaga, near Leon de Nicaragua,
therefore rather out of the direction before assigned. This important volcano,
which emits sulphureous vapours from a crater 300 feet deep, was ascended a
few years ago by my friend, Professor Julius Probel : he found the lava
composed of glassy felspar and augite. (Squier, vol. ii. p. 115 — 117.) On
the summit, 3517 feet high, there is a crater in which the vapours deposit
great masses of sulphur. At the foot of the volcano there is a mud-spring
(Salse ?).

El Viejo*, the northernmost of the crowded range of six volcanoes. It was
ascended and measured by Captain Sir Edward Belcher in 1 838. The result of
his measurement was 5560 feet : a later measurement by Squier gave 6000 feet.

NOTES. XClil

This volcano, which was very active in Dampier's time, is still burning. The
eruptions of burning scoriae are often seen in the town of Leon.

Quanacaure : rather to the north, out of the line from El Nuevo to El Viejo,
only twelve miles from the coast of the Gulf of Fonseca.

Consequina*, on the promontory which runs out at the southern end of the
Gulf of Fonseca (N. lat. 12° 50') ; celebrated for the dreadful eruption, preceded
by earthquake, 23rd January 1835. The great darkness produced by the fall
of ashes, similar to that which has sometimes been occasioned by Pichincha,
lasted forty-three hours. At a few feet distant torches could not be perceived,
respiration was impeded, and subterranean noises, resembling the discharge of
heavy ordnance, were heard not only in Balize, on the peninsula of Yucatan, but
also in Jamaica and on the high plain of Bogota, the latter being more than
8500 feet above the sea, and almost 560 geographical miles distant. (Juan
Galindo in Silliman's American Journal, vol. xxvjii.. 1835, p. 332—336; Acosta,
Viajes a los Andes, 1849, p. 56; and Squier, vol. ii. p. 110 — 113; drawings, p.
163 and 165.) Darwin (Journal of [Researches during the Voyage of the Beagle,
chap. 14, p. 291) calls attention to a remarkable coincidence of phsenomena :
after a long slumber the volcanoes of Conseguina in Central America, of Acon-
cagua and Corcovado (S. lat. 32f° and 43^°) in Chili* broke out on the same
day. Was this an accident ?

Volcano of Conchagua, or of Amalapa, at the northern entrance of the Gulf of
Fonseca, opposite to the volcano of Conseguina, at the fine Puerto de la Union,
the port of the neighbouring town of San Miguel.

From the State of Costa Rica to the volcano of Conchagua, the range of
twenty volcanoes follows a S.E. — N.W. direction, but at Conchagua, where it
enters the State of San Salvador, which, in a length of 160 geographical miles,
has five volcanoes, all still more or less active, the line of volcanoes, like the coast
line towards the Pacific, turns more E.S.E. — W.N.W. or even almost E.— W. ;
while on the eastern (Atlantic) side, about Cape Gracias d Dios, the coast-line
swells out suddenly in the territories of Honduras and Los Mosquitos. (Com-
pare text, p. 263.) It is not, as there remarked, until the lofty volcanoes
of Old Guatemala are passed, about the Laguna of Atitlan, that the former
general direction, N. 45° W., returns, and continues to Cbiapa and the isthmus
of Tehuantepec, when the abnormal E. — W. direction reappears in non-volcanic
chains. The remaining four volcanoes in the State of San Salvador, besides
Conchagua, are :

The volcano of San Miguel Bosotlan* (N. lat. 13° 35'), near the town of that
name ; the finest and most regular trachytic cone, next to the island-volcano
Ometepec, in the lake of Nicaragua. (Squier, vol. ii. p. 196.) The volcanic

XC17 NOTES.

forces are very active in Bosotlan : it had a great outpouring of lava on the 20th
of July 1844.

Volcano of San Vicente*, west of Rio de Lempa, between the towns of Sacate-
coluca and Sacatelepe. A great eruption of ashes took place, according to
Juarros, in 1643; and in January 1835, a very destructive earthquake accom-
panied an eruption of long duration.

Volcano of San Salvador (N. lat. 1 3° 47'), near the town of the same name.
The latest eruption took place in 1656. The whole district around is exposed
to severe earthquake-shocks; that of the 16th of April 1854, which was not
preceded by any noise, overthrew almost all the buildings in San Salvador.

Volcano of Izalco*, near the village of the same name ; often produces mu-
riate of ammonia. The first historically known eruption took place on the 23rd
of February 1770; subsequent eruptions, of which the light was seen at great
distances, took place in April 1798, 1805 to 1807, and 1825. (See above
p. 255 — 256, and Thompson, Official Visit to Guatemala, 1829, p. 512.)

Volcan de Pacaya* (N. lat. 14° 23'), about twelve miles south-east of the
town of New Guatemala, on the little Alpine lake Amatitlan ; a very active,
often flaming, volcano ; an extended ridge with three rounded summits. It had
great eruptions in 1565,*1651, 1671, 1677, and 1775; the last of these, in
which much lava flowed, was described by Juarros, who himself witnessed it.

Next follow the two volcanoes of Old Guatemala, named De Agua and De
Fuego ; in 14° 12' N. lat., near the coast.

Volcan de Agua : a trachytic cone near Escuintla, higher than the Peak of
Teneriffe ; surrounded by masses of obsidian (evidences of ancient eruptions ?).
This volcano, which enters the region of perpetual snow, received its name from
a great inundation which was attributed to it, and which, in September 1541,
destroyed the first city of Guatemala, and occasioned the foundation of the second,
now called Antigua Guatemala, which was built more to the N.N.W. (Was the
inundation produced -by earthquake and melting of snow ?)

Volcan de Fuego*, near Acatenango, twenty miles W.N.W. from the Volcan
de Agua. For their relative situation see the rare map (engraved at Guatemala
and given to me from thence) of the Alcalde mayor, Don Jose' Rossi y Rubi,
entitled, Bosqufjo del Espacio que media entre los estremos de la Provincia de
Suchitepeques y la Capital de Guatemala, 1800. The Volcan de Fuego is still
active, though much less so than formerly. The earlier great eruptions took
place in 1581, 1586, 1623, 1705, 1710, 1717, 1732, 1737, and 1799 ; but it
was not so much these as the destructive earthquakes which accompanied them,
that induced the Spanish Government, in the second half of the last century, to
forsake the second site, where the ruins of La Antigua Guatemala now stand,

NOTES. XCV

ana to oblige the inhabitants to establish themselves more to the north,
in the new town of Santiago de Guatemala. On this, as on the removal of the
site of Riobamba and other towns near the volcanoes of the Andes, there was
much dogmatic and passionate debate respecting the choice of a locality,
which might be supposed from previous experience to be little exposed to the
effects of neighbouring volcanoes, streams of lava, eruptions of scorise, and earth-
quakes ! The Volcan de Fuego had a great eruption in 1852, when it sent
forth a stream of lava towards the shore of the Pacific. Captain Basil Hall mea-
sured, under sail, the two volcanoes of Old Guatemala, and found for the Volcan
de Fuego 14,665, and for the Volcan de Agua 14,903 feet. Poggendorff ex-
amined the data on which these results were based, and inferred a less height,
i. e. about 13,110 feet for the mean of the two.

Volcan de Quesaltenango* (N. lat. 15° 10'), burning since 1821, and giving
out smoke : near the town of the same name: the three conical mounts which
bound the Alpine lake, Atitlan, on the south, are also supposed to be burning.
The volcano of Tajamulco named by Juarros cannot well be identical with the
volcano of Quesaltenango, as the latter is forty geographical miles to the N.W.
of the village of Tajamulco, south of Tejutla.

What are the two volcanoes of Sacatepeques and Sapotitlan named by Funel ;
or Brue"s Volcan de Amilpas ?

The great Volcano of Soconusco : near the boundary of Chiapa, twenty-eight
miles south of Ciudad Eeal, in N. lat. 16° 2'.

At the conclusion of this long note I must again recall that the barometric de-
terminations of height are partly given by Espinache, and partly taken from
the writings and maps of Daily, Squier, and Molina, and that they are expressed
in Paris feet (which in the translation have been converted into English feet).

("^ p. 265. — Of the volcanoes cited by me as active either formerly or at
present, the following eighteen, almost half the number, may be regarded as still
more or less so : Irasu and Turrialva, near Cartago ; el Rincon de la Vieja ;
Votos (?) and Orosi ; the island-volcano Ometepec, Nindiri, Momotombo, el
Nuevo (at the foot of the trachytic mountain las Pilas), Telica, el Viejo, Conse-
guina, San Miguel Bosotlan, San Vicente, Izalco, Pacaya, Volcan de Fuego (de
Guatemala), and Quesaltenango. The most recent eruptions were those of el
Nuevo, 18th April 1850; San Miguel Bosotlan, 1848; Conseguina and Sail
Vicente, 1835 ; Izalco, 1825 ; Volcan de Fuego, 1799 and 1852 ; and Pacaya
1775.

(392) p. 266. — Compare Squier, Nicaragua, vol. ii. p. 103, with p. 106 and
111, as well as his earlier short Notice on the Volcanoes of Central America,
1850, p. 7; and L. de Buch, lies Canaries, p. 506; where mention is made of the

XCvi X^/^o y J NOTES.

stream of lava which issued from Nindiri in 1775, and which has recently been
seen again by a very scientific traveller, Dr. Oersted.

(39S) p. 268.— See all the bases on which these determinations of position in
Mexico are founded, and their comparison with the observations of Don Joaquin
Ferrer, in my Recueil d'Observ. Astron. vol. ii. p. 521, 529, and 536 — 550, and
Essai Pol. sur la Nouv. Espagne, t. i. p. 55—59 and 176, t. ii. p. 173. Re-
specting the position of the Volcano of Colima I had myself early expressed
doubt (Essai Pol. t. i. p. 68, and t. ii. p. 180). By altitudes taken by Cap-
tain Basil Hall under sail, that volcano would be in N. lat. 19° 36', or half a
degree north of the position which I had inferred from the Itineraries ; but
without absolute determinations for Selagua and Petatlan, which I used as
points of departure. The latitude and the elevation (12,003 English feet), given
iu the text, are taken from Admiral Beechey (Voyage, Pt. II. p. 587). The
most recent map by Laurie (the Mexican and Central States of America) gives
19° 20' for the latitude. The latitude of Jorullo may also be two ( or three
minutes wrong, as when there I was much engaged with geological and topo-
graphical examinations, and neither sun nor stars were visible for the determina-
tion of latitude. Compare Basil Hall, Journal written on the coast of Chili, Peru,
and Mexico, 1824, vol. ii. p. 579 ; Beechey, Voyage, Pt. II. p. 587; and Hum-
boldt, Essai Pol. t. i. p. 68, t. ii. p. 180. In the faithful and highly picturesque
views which Moritz Rugendas has taken of the Volcano of Colima, and which are
preserved in the Berlin Museum, one may distinguish two mountains near each
other, one the proper volcano from which smoke always issues, which has but
little snow, and the other, the loftier Nevada, which enters deep into the region
of perpetual snow.

C394) P* 272. — The following is the result of the determinations of latitude
and longitude of the five linear groups of volcanoes in the Andes, and also a state-
ment of the distances between the groups, elucidating the relative amount of the
volcanic and non volcanic spaces : —

I. Group of Mexican volcanoes. The fissure on which they rise runs
east and west from Orizaba to Colima, for a distance of 392 geographical
miles, between 19° and 19° 20' N. lat. The Volcano of Turtla stands
apart, 128 miles to the east of Orizaba, near the coast of the Gulf of
Mexico, in a parallel (18° 28') half a degree more to the south.

II. Distance of the Mexican from the next, the Central American group
(being the distance from the Volcano of Orizaba to that of Soconusco in the
E.S.E.— W.N.W. direction) : 300 miles.

III. Group of Central American volcanoes: its length from S.E. to N.W.,
from Soconusco to Turrialva in Costa Rica, above 680 miles.

NOTES. XCV11

i

IV. Distance of the Central- American group from that of New Granada

and Quito : 628 miles.

V. Group of New Granada and Quito ; its length from the eruption in
the Paramo de Ruiz, north of the Volcan de Toltma, to the Volcano of
Sangay: 472 miles. The part of the chain of the Andes between the
Volcano of Purace' near Popayan and the southern part of the volcanic
mountain-knot of Pasto runs N.N.E. — S.S.W. Far to the east of the volca-
noes of Popayan, at the sources of the Rio Fragua, there is a very isolated
volcano which I have entered in my general map of the mountain-knots of
the South-American Cordilleras from the accounts communicated to me by
the missionaries of Timana ; distance from tho sea-shore, above 152 miles.

VI. Distance of the New Granada and Quito group from that of Peru and
Bolivia : 960 miles ; which is the greatest distance without volcanoes.

VII. Group of Peru and Bolivia, from the Volcan de Chacani and Are-
quipa to the Volcano of Atacama (16£° — 2li° S. lat.) : 420 miles.

VIII. Distance of the group of Peru and Bolivia from the Chilian volcanic
group: 540 miles. From the part of the desert of Atacama, at the edge of
which rises the Volcano of San Pedro, to far beyond Copiapo, and even to
the Volcano of Coquimbo (30° 5' S. lat.) in the long Cordillera west of the
two provinces of Catamarca and Rioja, there is not a single volcanic cone.

IX. Chilian group: from the Volcano of Coquimbo to that of San
Clemente, 968 miles.

These estimations of the length of the Cordilleras, with the curvatures arising
from changes of direction in the axis, from the parallel of the Mexican volcanoes
(in 19{° N. lat.), to the Volcano of San Clemente in Chili (46° 8' S. lat.), give
for an entire distance of 4968 miles a space of 2,540 miles occupied by five
linear groups of volcanoes (Mexico, Central America, New Granada and Quito,
Peru and Bolivia, and Chili), and a space of 2428 miles probably without vol-
canoes. The spaces are therefore nearly equal. I have given very definite nu-
merical relations, such as could be derived from a careful discussion of my own
and other maps and materials, in the hope of inducing others to further rectifi-
cation and improvement.

(s>5) p 272. — The group of Mexican volcanoes comprises those of Orizaba *,
Popocatepetl*, Toluca (or Cerro de San Miguel de Tutucuitlapilco), Jorullo*,
Colima*, and Turtla*. The still burning volcanoes are here, as in other similar
lists, marked by an asterisk.

(398^ p 272.— The volcanoes of the Central-American group are enumerated
in Notes 390 and 391.

C87) p. 272. — The group of New Granada and Quito comprises the Paramo
VOL. IV g

XCV111 NOTES.

y Volcan de Ruiz *, the volcanoes of Tolima, Purace' *, and Sotard near Popayan;
the Volcan del RioFragua (a river which falls into the Caqueta) ; the Volcanoes
of Pasto, el Azufral*, Cumbal*, Tuquerres*, Chiles, Imbaburu, Cotocachi,
Rucu-Pichincha, Antisana ( ?), Cotopaxi * Tungurahua *, Capac-Urcu or Altar
de los Collanes (?), and Sangay *

(398) p. 272. — The group of Southern Peru and Bolivia comprises the following
fourteen volcanoes, proceeding from North to South : —

Volcano of Chacani (according to Curzon and Meyen also called Char-
cani), visible from the town of Arequipa, on the right bank of the Rio
Quilca; according to Pentland, the most accurate geological explorer of
this region, in S. lat. 16° 11 ; 32 miles south of the Nevado of Chuqui-
bamba, estimated to exceed 19,000 feet in height. Manuscript Notices in
my possession give to the Volcano of Chacani 19,600 feet. On the south-
eastern part of the summit Curzon saw a large crater.

Volcano of Arequipa* : S. lat. 16° 20'; twelve miles north-east of the
town of the same name. Respecting its height (18,877 feet) see above,
p. 248, and Note 369. Thaddaeus Hanke, the botanist of Malaspina's ex-
pedition (1796), Samuel Curzon of the United States of North America
(1811), and Dr. Weddell (1847) ascended the summit, Meyen, in Au-
gust 1831, saw great columns of smoke rise from it ; a year before it had
erupted scoriae, but never lava. (Meyen's Reise urn die Erde, Th. II.
S. 33.)

Volcan de Omato : S. lat. 16° 50' ; it had a violent eruption in 1667.

Volcan de Uvillas, or Uvinas : south of Apo ; its latest eruptions were
in the 1 6th century.

Volcan de Pichu-Pichu : sixteen miles east of the town of Arequipa
(lat. 16° 25' S.), not far from the pass of Cangallo, 9,673 feet above the
sea.

Volcan Viejo : 16° 55' S. ; an enormous crater with lava streams and
much pumice.

The six last-named volcanoes compose the group of Arequipa.

Volcan de Tacora, or Chipicani, according to Pentland's fine Map of the
Lake of Titicaca : 17° 45' S., height 19,735 feet.

Volcan de Sahama* : 22,350 feet high ; 18° 7' S. ; a truncated cone
of the most regular form; see above, Note 371. Sahama, according to
Pentland, is 870 French feet higher than Chimborazo, but 6240 French
feet lower than Mount Everest in the Himalaya, which is now regarded as
the highest known summit on the globe. According to the last official
accounts from Colonel Waugh, 1st March 1856, the four highest moun-

NOTES. XC1X

tains in the Himalaya are : Mount Everest, or Gaurischanka, north-east of
Katmandoo, 29,000 feet; Kinchinjinga, north of Darjeeling, 28,154 feet ;
Dhawalagiri, 26,825 feet ; and Tschumalari (or Chamalari), 23,930 feet.
Volcano Pomarape, 21,700 feet, 18° 8' S., almost a twin mountain to
the next volcano.

Volcano Parinacota, 22,033 feet, 18° 12' S.

The group of four trachytic cones : Sahama, Pomarape, Parinacota, and
Gualatieri, which lies between the parallels of 18° 7' and 18° 25, is, ac-
cording to Pentland, higher than Chimborazo, higher than 21,424 feet.

Volcano Gualatieri* : 21,960 feet, 18° 25' S.; in the Bolivian province
Carangas ; according to Pentland burning strongly. (Hertha, Bd. xiii.
1829, S. 21.)

Not far from the Sahama group, in 18° 7' to 18° 25' S., the direction
both of the series of volcanoes and of the whole chain of the Andes to the
west of it, suddenly alters from N.W. — S.E. to N. — S., which continues to
be the general direction to the Straits of Magellan. I have already spoken
(Kosmos, Bd. i. S. 310 and 472 ; English edition, p. 284 and Note 347)
of this important inflection which is analogous to the one in the Bight of
Biafra on the west coast of Africa.

Volcano Isluga : 19° 20' S., in the province of Tarapaca, west of Ca-
rangas.

Volcan de San Pedro de Atacama : on the north-eastern edge of the
desert of the same name, according to the recent special map of the water-
less sandy waste (desierto) of Atacama by Dr. Philippi ; lat. 22° 16' S.,
sixteen geographical miles north-east of the little town of San Pedro, not
far from the great Nevado de Chorolque.

There is no volcano from 213° to 30° S. ; and after this long interruption of
568 miles, volcanic activity first reappears in the Volcano of Coquimbo. For
the existence of a Volcano of Copiapo in lat. 27° 28' is denied by Meyen, though
believed by Philippi who knew the country well.

(399) p. 272. — Our geographical and geological knowledge of the Chilian
group, or line of volcanoes, is largely indebted to the able investigations of Cap-
tain Fitz-Roy and of Charles Darwin, in the memorable Expedition of the " Ad -
venture " and " Beagle." Darwin embraced, with the comprehensive glance and
spirit of generalisation which belong to him, the connection in one point of view
of earthquake phenomena and volcanic eruptions. The great natural event which,
on the 22nd of November 1822, destroyed the town of Copiapo, was accom-
panied by the elevation of a considerable strip of land along the coast ; and
during the very similar phenomenon which, on the 20th of February 1835, so

C NOTES.

greatly injured the town of Concepcion, a submarine volcano broke forth near
Bacalao Head, on the Island of Chiloe, with a fiery eruption which raged vio-
lently for a day and a half. All this, with previous similar occurrences, under
similar conditions, confirms the belief, that the range of rocky islets which, south
of Valdivia and Fuerte Maullin, lie opposite to the fiords of the mainland ; viz.
Chiloe, the Archipelago of the Chonos and Huaytecas, the Peninsula de Tres
Monies, and the Islas de la Campana, de la Madre de Dios, de Santa Lucia,
and los Lobos, from 39° 53' S. lat. to the entrance of Magellan Strait in
52° 16', are the jagged crest of a sunken westernmost Cordillera. It is true that
we see no open trachytic cone, no volcano, in these fractis ex sequore terris, but
single submarine eruptions sometimes following, sometimes preceding, great
earthquakes, appear to point to the existence of this western fissure. (Dar-
win, on the Connexion of Volcanic Phsenomena, the Formation of Mountain
Chains, and the Effect of the same Powers, by which Continents are elevated :
in the Transactions of the Geological Society, second series, vol. v. Pt. III. 1840,
p. 606—615 and 629 — 631 ; Hmnboldt, Essai Pol. sur la Nouv. Espagne,
t. i. p. 190, and t. iv. p. 287.)

The twenty-four volcanoes of the Chilian group, proceeding from north to
south, from the parallel of Coquimbo to 46° S. lat., are as follows : —
a. Between the parallels of Coquimbo and Valparaiso :

Volcan de Coquimbo (3&° 5' S.) ; Meyen, Th. 1, S. 385.
„ Limari.
„ Chuapri.

„ Aconcagua*: W.N.W. of Mendoza, 32° 39' S. ; height
23,004 feet, according to Kellet (see above, Note 371) ; but, according
to the most recent trigonometric measurement of the engineer Amado
Pissis (1854), only 22,301 feet, lower therefore than Sahama, to which
Pentland now assigns 22,350 feet ; Gilliss, U. S. Naval Astr. Exp. to
Chili, voL i. p. 13" : Pissis has given in full the geodetical elements of
his measurement (which required eight triangles) in the Anales de la Uni-
versidad de Chile, 1852, p, 219.

Peak Tupungato has assigned to it by Gilliss 22,450 feet, and 33° 22'
S. ; but in the Map of the Province of Santiago, by Pissis (Gilliss, p. 45),
22,016 feet are given. The latter value (6710 metres) is retained by
Pissis in the Anales de Chile, 1850, p. 12.
b. Between the parallels of Valparaiso and Concepcion :

Maypu* : according to Gilliss (vol. i. p. 13) in 34° 17' S. lat. (but in
his General Map of Chili, no doubt erroneously, in 33° 47'), and the height
1 7,662 feet ; ascended by Meyen. The trachytic rock of the summit has

NOTES. Cl

broken through Upper Jurassic strata in which von Buch recognised Exo-
gyra Couloni, Trigonia costata, and Ammonites biplex, taken from heights
of 9600 feet. (Description Physique des lies Canaries, 1836, p. 471.)
No streams of lava, but flames and eruptions of scoriae from the crater.

Peteroa * : east of Talca, 34° 53' S. ; often burning ; and, according to
Molina's description, having had a great eruption Dec. 3, 1762 ; was vi-
sited, in 1831, by the able explorer of nature, Gay.

Volcan de Chilian : 36° 2' S. ; the district has been described by the
Missionary Havestadt of Munster. Near it is the Nevado Descabezado
(35° 1' S.), which Domeyko ascended, and Molina (erroneously) stated to
be the highest mountain in Chili. Gilliss estimates its height at 13,100
feet. (U. S. Naval Astr. Exp. 1855, vol. i. p. 16 and 371.)

Tucapel : west of the town of Concepcion ; also called Silla Veluda ;
perhaps an unopened trachytic mount connected with the burning volcano
of Antuco.
c. Between the parallels of Concepcion and Valdivia :

Antuco * : 37° 7' S. ; geologically described in detail by Pb'ppig, " a
basaltic crater of elevation, out of which the trachytic cone rises ; lava
streams burst forth at the foot of the cone, and more rarely from the
summit crater." (Poppig, Reise in Chile und Peru, Bd. i. S. 364.)
One of these streams was still flowing in 1828. In 1845, the diligent ex-
plorer Domeyko found it in full activity ; he gave its height only 8918
feet. (Pentland, in Mary Somerville's Phys. Geography, vol. i. p. 186.)
Gilliss gives for its height 9242 feet, and mentions new eruptions in 1853.
According to an account communicated to me by the last-named distin-
guished American astronomer it would appear that, in the interior of the
Cordillera, on the 25th of Nov. 1847, a new volcano was upheaved forming
a Hill* 320 feet high. The sulphurous and fiery eruptions were seen by
Domeyko for more than a year. Far to the east of the Volcano Antuco,
in a parallel chain of the Andes, Poppig mentions two other active volca-
noes : Punhamuidda * and Unalavquen.*

Volcano Callaqui.

Volcan de Villarica* : 39° 14' S.

Volcano Chinal : 39° 35' S.

Volcan de Panguipulli * : according to Major Philippi in 40f° S.
d. Between the parallels of Valdivia and the southernmost cape of the Island
of Chiloe :

Volcano Ranco.

„ Osorno, or Llanquihue : 41° 9' S. : height, 7442 feet.

Cll NOTES.

Volcan de Calbuco* : 41° 12' S.
Volcano Guanahuca (Guanegue ?).

Volcano Minchinmadom : 42° 48' S. ; height, about 8000.
Volcan del Corcovado* : 43° 12' S., height, 7509 feet.
Volcano Yanteles (Yntales) : 43° 29' S. ; height, 8029 feet. On the
four last-mentioned heights, see Fitz-Roy (Exped. of the " Beagle," vol. iii.
p. 275), and Gilliss, vol. i. p. 13.

Volcano San Clemente: opposite to the Peninsula de Tres Montes, which,
according to Darwin, consists of granite; 46° 8' S. On the great Map of
South America by La Cruz, a more southern volcano, called de los Gigan-
tes,'is marked opposite to the islands de la Madre de Dios, in 51° 4' S.
Its existence is very doubtful.

The latitudes in the above Table are for the most part taken from the Maps
of Pissis, Allan Campbell, and Claude Gay, in the excellent work of Gilliss
(1855).

(40°) p. 274.— Humboldt, Kleinere Schriften, Bd. i. S. 90.

(401) p. 274. — 24 January 1804. See my Essai Pol. sur la Nouv. Espagne,
t. i. p. 166.

(402) p. 277.— The mica-schist of the mountain-knot de los Robles, lat. 2° 2',
and of the Paramo de las Papas (lat. 2° 20'), contains two mountain-lakes,
Laguna de S. Jago and Laguna del Buey, less than six miles from each other,
from the first-named of which flows the Cauca, and from the second the Mag-
dalena River. These rivers are soon after divided from each other by a central
mountain- chain, and reunite, in lat. 9° 27', in the plains of Mombox and Tene-
rife. The above-mentioned knot, between Popayan, Almaguer, and Timana, is
of great importance in regard to the geological question whether the Andes of
Chili, Peru, Bolivia, Quito, and New Granada are connected with the chain of
the Panama Isthmus, and thus with that of Veragua and the volcanic ranges of
Costa Rica and the whole of Central America. In my maps of 1816, 1827, and
1831 (the "mountain -systems" of which have been more widely circulated by
Brud in Joaquin Acosta's fine Map of New Granada (1847), and other maps),
I have shown the manner in which, in N. lat. 2° 1 0', the chain of the Andes
trifurcates, the western cordillera running between the valley of the Rio Cauca
and the Rio Atrato, the central one between the Cauca and the Magdalena, and
the eastern between the Magdalena and the plains, or llanos, which are watered
by the tributaries of the Amazons and the Orinoco. I was able to show the
special direction of each of these Cordilleras from a great number of points which
are part of my series of astronomical determinations of places, of which in South
America alone I obtained 1 52 by star culminations

NOTES. Cill

The Western Cordillera runs east of RioDagua, west of Cazeres Roldanilla, Toro
and Anserma near Cartago, S.S.W.— N.N.E. to the Salto de San Antonio in the
Kio Cauca (lat. 5° 14'), which is south-west of the Vega de Supia. From thence
and up to the Alto del Viento, 9600 feet high (Cordillera de Abibe, or' Avidi,
lat. 7° 12'), the chain increases considerably in height and extent, and in the Pro-
vince of Antioquia unites with the middle or central cordillera. Farther to the
north, towards the sources of the Rios Lucio and Guacuba, the chain sinks and
subdivides into ranges of hills. The Cordillera Occidental, which, at the mouth
cf the Dagua in the Bahia de San Buenaventura, is only about thirty miles from
the coast line (lat. 3° 50'), is twice as far from it in the parallel of Quibdo in
the Choco (lat. 5° 48 ). This remark has some importance, because we must
not confound with the western chain of the Andes the high hilly land and the chain
of hills which in this province, rich in gold-washings, runs from Novita and
Tado along the right bank of the Rio San Juan, and the left bank of the Rio
Atrato, from south to north. It is this inconsiderable range of hills which, in
the Quebrada de la Raspadura, is traversed by the Canal of the Monk, which con-
nects two rivers (the Rio San Juan, or Noanama, and the Rio Quibdo, an affluent
of the Atrato), and thereby brings the two oceans into connection (Humboldt,
Essai Pol. t. i. p. 235) ; and it is also this range which was seen in Captain
Kellet's instructive expedition, between the Bahia de Cupica, in lat. 6° 42', so
long fruitlessly extolled by me, and the sources of the Napipi, which falls into the
Atrato. (Compare Essai Pol. t. i. p. 231 ; and Robert Fitz-Roy's Considerations
on the great Isthmus of Central America, in the Journal of the Royal Geogr.
Soc. vol. xx. 1851, p. 178, 180, and 186.)

The middle chain of the Andes (Cordillera Central), persistently the highest,
entering the region of perpetual snow, and running throughout its whole extent,
like the western chain, almost north and south, begins about thirty-four miles
north-east of Popayan with the Paramos of Quanacos, Huila, Iraca, and Chinche.
Further on towards the north rise, between Buga and Chaparral, the long
extended ridges of the Nevadode Baraguan (lat. 4° 11'), la Montana de Quindio,
the snow-covered truncated cone of Tolima, the Volcano and Paramo deRuiz, and
the Mesa de Herveo. These high and rude mountain solitudes, designated in the
Spanish language by the name of Paramos, are distinguished by temperature and
by a peculiar character of vegetation, and in the part of the tropics which I am
now describing, occupy elevations which, by many measurements, may be stated to
average from 9500 to 11,000 French feet (10,124 to 11,723 Engl.). In the
parallel of Mariquita. the Herveo, and the Salto de San Antonio, begins the mas-
sive junction, or coalition, of the western and central chain which has been al-
ready alluded to. This coalition is most striking between the Salto de San

Civ NOTES.

Antonio and the Angostura and Cascada de Caramanta near Supia. There
lies the highland and the difficultly accessible Province of Antioquia which, ac-
cording to Manuel Restrepo, extends from 5|° to 8f° N. lat., and in which,
still proceeding from south to north, we name successively : Arma, Sonson ,
north of the sources of the Rio Samana ; Marinilla, Rio Negro (6842 feet), and
Medellin (4847 feet) ; the plateau of Santa Rosa (8466 feet) ; and the Valle de
Osos. Beyond Cazeres and Zaragoza, towards the confluence of the Cauca and
the Nechi, the mountain-chain properly so called disappears ; and the eastern de-
clivity of the Cerros de San Lucar which, when navigating and surveying the
Magdalena River, I saw from Badillas, in 8° 1' N., and Paturia, 7° 36' N., is
only rendered sensible by the contrast with the broad plain of the river.

The Eastern Cordillera offers the geographical interest not only of dividing the
whole northern mountain-system of New Granada from the lowlands, from which
the waters flow partly through the Caguan and Caqueta to the Amazons, and
partly through the Guaviare, Meta, and Apure to the Orinoco ; but also of being
decidedly connected with the coast-chain of Caracas. There is here a conjunc-
tion of mountain-ridges which have been elevated over two fissures of very dif-
ferent directions, and probably also at very different times. The eastern cor-
dillera departs much more than the other two from the direction of the meridian,
deviating so much towards N.E. that at the snow-covered mountains of Merida,
in 8° 10' N. lat., it is already 5° more easterly than at the mountain-knot de los
Robles not far from the Ceja and Timana. North of the Paramo de la Suma
Paz, east of la Purificacion, on the western declivity of the Paramo of Chinzaga,
at a height of only 8760 feet, there rises above the oak forest the fine but tree-
less high plain of Bogota (lat. 4° 36'). It extends over about eighteen (German)
geographical square miles, and its situation presents a striking similarity to that
of the basin of Kashmeer (which is however, according to Victor Jacquemont,
3410 feet lower, and belongs to the south-western declivity of the Himalayan
chain). Next to the plateau of Bogota and the Paramo de Chinzaga there follow
in the eastern cordillera of the Andes, towards the north-east, the Paramos of
Guachaneque, above Tunja; of Zoraca, above Sogamoso; of Chita (16,000 feet ?)
near the sources of the Rio Casanare, a tributary of the Meta; of the Almorzadero
(12,854 feet) near Socorro ; of Cacota (10,986 feet) near Pamplona ; and of
Laura and Porquera near la Grita. Here, between Pamplona, Salazar, and
Rosario (between 7° 8' and 7° 50' N. lat.), is the small mountain-knot, from
whence a crest stretches northward towards Ocana and Valle de Upar, west
of the Laguna de Macaraibo, and unites with the advanced posts of the Sierra
Nevada de Santa Marta (19,000 feet ?). The higher and more considerable
crest continues to run in the previous N.E. direction towards Merida, Truxillo,

NOTES. CV

and Barquisimeto, there to unite itself with the granitic coast-chain of Vene-
zuela, east of the Laguna de Macaraibo and west of Puerto Cabello. From the
Grita and the Paramo de Porquera, the Eastern Cordillera rises again to an ex-
traordinary height. There follow, between the parallels of 8° 5' and 9° 7', the
Sierra Nevada de Merida (Mucuchies), explored by Boussingault, and of which
Codazzi has assigned the height, by trigonometrical measurement, 15,165 feet,
and the four Paramos of Timotes, Niquitao, Bocond, and de las Kosas, full of the
most beautiful alpine plants. (Compare Codazzi, Kesumen de la Geografia de
Venezuela, 1841, p. 12 and 495 ; also rny Asie Centrale, on the height of per-
petual snow in this zone, t. iii. p. 258 — 262.) The western cordillera is en-
tirely without volcanic activity ; in the middle one it prevails up to Tolima and
the Paramo de Euiz, which, however, are almost three degrees of latitude from
the Volcan de Purace'. The eastern cordillera has near its eastern declivity,
at the origin of the Rio Fragua, north-east of Mocoa and south-east of Ti-
mana, a smoking hill : more distant from the shore of the Pacific than any other
still active volcano in the New Continent. An exact knowledge of the local
relations of the volcanoes to the ramifications of the mountain-chains is highly
important towards the completion of the geology of volcanoes. All older maps,
excepting that of the highland of Quito, are only calculated to mislead.

(403) p. 277. — Pentland, in Mary Somerville's Phys. Geography, 1851, vol. i.
p. 185. The Peak of Vilcanoto (17,020 feet, in 14° 28' lat ), a part of the
mass of mountains of that name, and directed east and west, forms the northern
extremity of the high plain in which the Lake of Titicaca, 88 miles long, appears
only a small inland sea

(404) p. 278. — Compare Darwin, Journal of Researches into the Natural His-
tory and Geology during the Voyage of the "Beagle," 1845, p. 275, 291, and
310.

(405) p. 280.— Junghuhn, Java, Bd. i. S. 79.

(i406) p. 280.— Junghuhn, Java, Bd. iii. S. 155 ; and Goppert, die Tertiarflora
auf der Insel Java nach den Entdeckungen von Junghuhn (1854), S. 17. The
absence of monocotyledons is, however, only to be understood of the silicified
trunks of trees found on the surface and particularly in the rivers of the Ban-
tam district ; in the subterrannean coal-beds remains of palms are found, be-
longing to the two genera of Flabellaria and Amesoneuron. See Goppert, S. 31
and 35.

(*") p. 281. — On the signification of the word Mera, and the conjectures
communicated to me by Burnouf respecting its connection with Mera (a Sans-
crit word for sea), see my Asie Centrale, t. i. p. 114 — 116; and Lassen's In-

CV1 NOTES.

dische Alterthumskunde, Bd. i. S. 847 ; he is inclined to regard the name as
not of Sanscrit origin.

(408) p. 281.— Kosmos, Bd. iv. S. 284, and Anm. 6 (English edition, p. 239,
and Note 330).

(409) p. 281. — Gunung is the Javanese word for mountain; in Malay, gunong.
It is remarkable that its use does not extend further over the enormous range
occupied by languages of the Malay stock. See the comparative table of words
in my brother's work, Ueber die Kawi Sprache, Bd. ii. S. 249, No. 62. As it is
the custom to prefix this word to the names of mountains in Java, it is indi-
cated in the text for the most part by a simple G.

(41°) p. 281.— Le'op. de Buch, Description Physique des lies Canaries, 1836,
p. 419. But it is not only Java (Jnnghuhn, Th. i. S. 61, and Th. ii. S. 547)
which has a mountain, G. Semeru, a little higher than the Peak of Teneriffe,
being 1 2,235 feet ; the also still active Indrapura in Sumatra, which would
seem to have been less exactly measured, is supposed 11,500 French feet
(12,256 Engl.) ;— (See Th. i. S. 78, and profile map, No. 1.) Next to Indra-
pura come, in Sumatra, Telaman one of the summits of Ophir (not 13,834,
but only 9603 feet high), and Merapi (according to Dr. Horner, 9570 feet) the
most active of the thirteen volcanoes in Sumatra, which, however (Th. ii. S. 294,
and Junghuhn's Battalander, 1847, Th. i. S. 25), must not be confounded with
two Javanese volcanoes of similar name : the celebrated Merapi near Jogjakerta
(9208 feet), and the Merapi which is the eastern part of the summit of the
Volcano Idjen (8595). The name Meru, combined with the Malayan and Ja-
panese word api, fire, has been supposed to be recognised in Merapi.

(4n) p. 282.— Junghuhn, Java, Bd. i. S. 80.

(412) p. 282. — Compare Joseph Hooker, Sketch Map of Sikhim, 1850, and
in his Himalaya Journals, vol. i. 1854, Map of part of Bengal ; as well as
Strachey, Map of West-Nari, in his Physical Geography of Western Thibet,
1853.

(41S) p. 283.— Junghuhn, Java, Bd. ii. fig. ix. S. 572, 596, and 601—604.
From 1829 to 1848, the small eruption-crater, Bromo, had eight fiery eruptions.
The crater-lake which had disappeared in 1842 had reformed in 1848 ; but
according to the observations of B. Van Herwerden, the presence of water had
not prevented the ejection of glowing scoriae, which were hurled to a distance.

(414) p. 283.— Junghuhn, Bd. ii. S. 624—641.

(415) p. 284.— Gunung Pepandajan was ascended by Reinwardtin 1819, and
by Junghuhn in 1837. Junghuhn, who examined carefully the parts round the
mountain, which are covered with numerous angular blocks of lava, and com-
pared them with all the earliest accounts he could obtain, looks upon the state-

NOTES. CV11

ment which has been circulated in many valuable writings, that during the erup-
tion of 1772 a part of the mountain had fallen in, and an adjacent area of
many square miles had sunk down, as greatly exaggerated. (Junghuhn, B. ii.
S. 98 and 100.)

(4US) p. 284.— Kosmos, Bd. iv. S. 495, Anm. 30 (Engl. edition, Note 254),
and Voyage aux Ke'gions e'quinox. t. ii. p. 16.

(417) p. 285.— Junghuhn, Bd. ii. S. 241—246.

(418) p. 286.— Junghuhn, Bd. ii. S. 566, 590, and 607—609.

(419) p. 286. — Leop. von Buch, Phys. Beschr. der Canarischen Inseln, S. 206,
218, 248, and 289.

(42°) p. 286. — Barranco and barranca, both terms having the same meaning,
and both in use in Spanish America, do indeed properly denote a water-furrow,
a water-rift : " la quiebra que hacen en la tierra las corrientes de las aguas ; " —
" una torrente que hace barrancas ; " they are, however, also employed for every
kind of ravine. I doubt whether it is correct to connect the word with barro,
" clay, soft moist mud, road mire."

(421) p. 287. — Lyell, Manual of Elementary Geology, 1855, chap.xxix. p. 497.
The most striking analogy with the phenomenon of the regular ribs in Java is
presented by the surface of the mantle of the Somma at Vesuvius, on the seventy
foldings of which much light has been thrown by an ingenious and exact observer,
Julius Schmidt (Die Eruption des Vesuvs im Mai 1855, S. 101 — 109). In the
view of Leopold von Buch these furrows were not originally made by the running
down of rain-water (are not therefore " fiumare ") but are rifts, as it were starred
cracks, produced in the first upheaval of the volcano. The generally radial
position of the lateral eruptions relatively to the axis of the volcano seems con-
nected with this (S. 129).

(4~) p. 287. — " L'obsidienne et par consequent les pierres-ponces sont aussi
rares a Java que le trachyte lui-meme. Un autre fait tres-curieux c'est 1'ab-
sence de toute coulee de lave dans cette ile volcanique. M. Rein ward t, qui lui-
meme a observe un grand nombre d'e'ruptions, dit expresse'ment qu'on n'a jamais
eu d'exemples que 1'eruption la plus violente et la plus de'vastatrice ait ete' ac-
compagne'e de laves." (Le'op. de Buch, Description des lies Canaries, p. 419.)
In the specimens of the volcanic rocks of Java for which the Cabinet of Minerals
at Berlin is indebted to Dr. Junghuhn, dioritic-trachytes, composed of oligo-
clase and hornblende, may be most clearly recognised at Burungagung, S. 255,
of the Leidner-Catalogue; at Tjinas, S. 232, and from G. Parang, situated in the
Batu-Gangi district. This is also identically the same formation as the dioritic-
trachytes of the volcanoes of Orizaba and Toluca in Mexico, of the Island of
Panaria, one of the Li pan Isles, and of ^Egina in the ^Egean Sea.

CVin NOTES.

C42*) p. 287.— Junghuhn, Bd. ii. S. 309 and 314. The fiery stripes which
•were seen on G. Merapi were formed by streams of glowing scoriae (" traine'es de
fragmens "), disconnected masses, which being erupted roll down on the same
side, and being of very different weights, overtake and strike against each other
in their rapid passage down the abrupt descent. In the eruption of G. Lamon-
gan, on the 26th of March 1847, such a stream divided into two branches some
hundred feet below its origin. Junghuhn says expressly (Bd. ii. S. 767), that
the fiery streaks consisted not of true molten lava, bat of crowded fragments
rolling down after each other. G. Lamongan and G. Semeru (the first, 5340,
and the second, 12,235 feet high) are, among the volcanoes of Java, those which,
by their activity during long periods, most nearly resemble Stromboli, which is
scarcely 2960 feet high. The ejections of scorise occurred in G. Lamongan (in
the eruptions of July 1838 and March 1847) after pauses of fifteen or twenty
minutes ; and in G. Semeru (in the eruptions of August 1836 and September
1844) after pauses of from one and a half to three hours. (Bd. ii. S. 554 and
765 — 769.) In Stromboli itself, besides numerous eruptions of scorise, there
are small and rare outpourings of lava which, being detained by obstacles, some-
times become hardened on the slope of the cone. I attach great importance to
the different forms of continuity (as in lavas) or of disconnection (as in detached
scoriae) of the wholly or half fused materials which are ejected or poured out,
either from the same or from different volcanoes. Analogous researches, con-
ducted under the guidance of " leading ideas" in various regions of the globe,
are much to be desired, in order to remedy the partial views which arise from
the contemplation being limited to the four European active volcanoes. The
question proposed by myself in 1802, and by my friend Boussingault in 1831,
whether the Volcano Antisana, in the Cordillera of Quito, had sent forth streams
of lava, which will be further touched upon, may perhaps find its solution in
peculiarities to which fluid lava streams are subject. The essential character
of a stream of lava is that of a uniform connected fluid, a band-like current,
from which, as it cools and solidifies, scales detach themselves at the surface.
These scales, beneath which the almost homogeneous lava long continues to
flow, become pushed up, either obliquely or vertically, by the inequality of the
internal movement and by the development of hot gases ; and thus, when
several lava streams have flowed together into a kind of lava lake, as in Iceland,
there is produced, after cooling, a " field of fragments." The Spaniards, espe-
cially, in Mexico, call such spaces, which are very inconvenient to pass over,
" Malpais." Such fields of lava fragments, which are often met with on the
plain at the foot of a volcano, remind us of a lake of which the frozen surface
has been broken up into blocks of ice.

NOTES. C1X

(424) p. 287. — Thex name G. Idjen may, according to Buschmann, be de-
rived from the Javanese word hidjen: single, alone ; itself a derivative from the
substantive hidji or widji : a "grain," or "seed," which combined with sa
expresses the number one. On the etymology of Tengger, see my brother's
comprehensive Notice, " Ueber die Verbindungen zwischen Java und Indien "
(Kawi-Sprache, Bd. i. S. 188), where the historic importance of the Tengger
mountains is shown as being inhabited by a small tribe who have retained their
ancient Indo-Javanese belief, in opposition to the Mahometanism which now pre-
vails generally over the island. Junghuhn, who is very diligent in explaining the
mountain names from the Kawi language, says (Th. ii. S. 554) that " tengger "
means " hill," as is also said in Gericke's Javanese Dictionary ( Javaansch-
Nederduitsch Woordenboek, Amst. 1847). Slamat, the name of the high vol-
cano of Tegal, is the known Arabic word selamat, which means good speed,
good success, hail.

(425) p. 287.— Junghuhn, Bd. ii. : Selamat, S. 153 and 163 ; Idjen, S. 698 ;
Tengger, S. 773.

(426) p. 287.— Bd. ii. S. 760—762.

(427) p. 289.— Atlas Ge'ographique et Physique, accompanying the Eel. Hist.
(1814) PL 28 and 29.

(428) p. 289.— Kosmos, Bd. iv. S. 311—313 (English edition, p. 267—269).

(429) p. 290. — Kosmos, Bd. i. S. 216 and 444 (English edition, p. 196 and
Note 187), Bd. iv. S. 226 (English edition, p. 177).

(43°) p. 291. — In my Essai Politique sur ia Nouvelle Espagne, in the two
editions of 1811 and 1827 (in the latter, t. ii. p. 165—175), I gave, as the
nature of the work required, only an abridged extract from my Journal, without the
topographical plan or map of elevations. From the importance attached to such
an event in the middle of the last century, I have thought it well to complete the
extract for the present occasion. I am indebted for additional particulars re-
specting the new Volcano of Jorullo to an official document (written only three
weeks after the day of the first eruption), which was first discovered in 1830 by
a scientifically well-informed Mexican ecclesiastic, Don Juan Jose' Pastor Morales.
Some of my information was derived from verbal communications made to me by
my companion Don Ramon Espelde, who had obtained the information himself
from surviving eye-witnesses of the event. Morales discovered in the archives
of the bishop of Michuacan a Report addressed by Joaquin de Ansogorri, priest
in the Indian village of la Guacana, to his bishop, dated 19 Oct. 1759. Ober-
bergrath Burkart, in his instructive work entitled Aufenthalt und Reisen in
Mexico (1336), has also given a short extract from it (Bd. i. S. 230). At the
time of my journey, Don Rumon Espelde was living on the plain of Jorullo, and

CX NOTES.

had the merit of having been the first to ascend to the summit of the volcano.
Some years later he joined the expedition made thither by the In tend en te Cor-
regidor Don Juan Antonio de Riano, on the 10th of March 1789, to which be-
longed also a well-informed German, Franz Fischer, who had entered the
Spanish service as commissary of mines. It was through Fischer that the name
of Jorullo was first mentioned in Germany, by a letter which was printed in the
Schriften der Gesellschaft der Bergbaukunde, Bd. ii. S. 441. But the rise of the
new volcano had been spoken of still earlier, in Italy, in Clavigero's Storia Antica
del Messico (Cesena, 1780, t. i. p. 42), and in the poetic work Rusticatio Mexi-
cana of Pater Raphael Landivar (ed. altera, Bologna, 1782, p. 17). Clavigero,
in his estimable work, erroneously places the rise of the new volcano (of which he
writes the name Juruyo) in the year 1760, and enlarges his description by ac-
counts of showers of ashes extending as far as Queretaro, which had been given
him in 1766 by Don Juan Manuel de Bustamante, governor of the Province of
\ralladolid de Michuacan, as himself an eye-witness of the phenomenon. Lan-
divar, who, like Ovid, was an enthusiastic adherent of our theory of upheaval,
in well-sounding hexameters makes the colossus rise to the full height of three
milliaria, and the thermal springs flow " cold by day and warm by night."
However, I myself saw the centigrade thermometer rise to 52|° (126°'5 Fahr.)
in the water of the Rio Cuitimba, at noon.

Antonio de Alcedo, in the 5th part of his great and usefal Diccionario Geo-
grafico-Historico de las Indias Occidentales 6 America, 1789, — therefore in the
same year in which the Report of Riano and Fischer appeared in the Gazeta de
Mexico, — gave in the article Xurallo (p. 374 — 375) the interesting notice, that
when the earthquakes began in the Play as (29 June 1759), the Volcano of
Colima which was then in eruption was suddenly quieted, "although it is seventy
leguas (as Alcedo says ; according to my map only 1 12 geogr. miles) from the
Playas." He adds, " It is supposed that the matter in the bowels of the earth
met with obstacles to its continuing its former course, and at the same time
found openings towards the east, and thus was diverted — para reventar en
Xurullo " Exact topographical data for the neighbourhood of the volcano are
also to be found in Juan Jose' Martinez de Lejarza's Analisis Estadistico de la
Provincia de Michuacan, en 1822 (Mexico, 1824, p. 125, 129, 130, and 131).
The statement of the Author, who resided not far from Jorullo, that since my
visit the mountain had shown no sign of increased volcanic activity, was first con-
tradicted by reports of a new outbreak in 1819. (Lyell, Principles of Geology,
1855, p. 430.) As the position of Jorullo in latitude is not without importance,
my attention has been drawn to the circumstance that Lejarza, who elsewhere
always follows my determinations of latitude and longitude, and gives the Ion-

NOTES. CXI

gitnde of Jorullo as I do, 2° 25' W. of Mexico (101° 29' W. of Greenwich), dif-
fers from me in the latitude. Is his latitude for Jorullo (18° 53' 30"), which
comes very near to that of Popocatepetl (18° 59' 47"), founded on later obser-
vations with which I am unacquainted ? In my Recueil d'Observations Astrono-
miques, vol. ii. p. 521, I have said expressly, "Latitude supposee, 19° 8', de-
rived from good star observations at Valladolid, which gave 19° 52' 8", and
from the Itinerary direction." I only subsequently recognised the importance of
the latitude of Jorullo when I was laying down the large Map of Mexico, and
entering upon it the volcanic chain which runs east and west.

As in these considerations on the origin of Jorullo I have repeatedly alluded
to traditionary stories still prevalent in the country, I will refer at the end of
this long note to what is still a very popular tale, which was alluded to
by me in another work (Essai Pol. sur la Nouv. Espagne, t. ii. 1827, p. 172):
"Selon la credulite des indigenes, ces changemens extraordinaires que nous

venons de de'crire sont 1'ouvrage des moines Aux Playas de Jorullo,

dans la chaumiere que nous habitions, notre hote indien nous raconta qu'en 1759
des Capucins en mission precherent k 1'habitation de San Pedro, mais que,
n'ayant pas trouve un accueil favorable, ils chargerent cette plaine, alors si belle
et si fertile, des imprecations les plus horribles et les plus complique'es ; ils pro-
phe'tiserent que d'abord 1'habitation seroit engloutie par des flammes qui sorti-
roient de la terre, et que plus tard 1'air ambiant se refroidiroit a tel point que
les montagnes voisines resteroient eternellement couvertes de neige et de glace.
La premiere de ces male'dictions ayant eu des suites si funestes, le has peuple
indien voit dejk dans le refroidissement progressif du volcan le pre'sage d'un
hiver perpetuel."

The first printed account of the catastrophe (next to the poetical one of Lan-
divar) was no doubt that already referred to in the Gazeta de Mexico, de 5 de
Mayo 1789 (t. iii. Num. 30, p. 293—297). It had the modest heading :
"Superficial y nada facultativa Descripcion del estado en que se hallaba el
Volcan de Jorullo, la manana del dia 10 de Marzo de 1789 ;" and was occa-
sioned by the expedition of Riano, Franz Fischer, and Espelde. Later, in 1791,
the botanist Mocifio and Don Martin Sesse, from the Nautico-Astronomical Ex-
pedition of Malaspina, visited Jorullo from the Pacific coast.

(431) p. 295. — My barometric measurements give: for Mexico, 7469 feet ;
Valladolid, 6407 feet ; Patzcuaro, 7225 feet ; Ario, 6353 feet ; Aguasarco,
4988 feet ; and for the old plain of the Playas de Jorullo, 2583 feet. (Humb.
Observ. Astron. vol. i. p. 327 ; Nivellement barome'trique, No. 367 — 370.)

C432) p. 295.— I find, taking the old level of the Playas at 404 toises (2583
feet): for the convexity of the Malpais, 3114 feet ; for the crest of the great

CX11 NOTES.

lava stream, 3837 feet ; for the highest margin of the crater, 4265 feet ; and
for the lowest point of the crater on which we could set up the barometer,
4118 feet. We thus obtained for the height of the summit of Jorullo above
the ancient plain 1682 feet.

(433) p. 296. — Burkart, Aufenthalt und Keisen in Mexico, in den Jahren
1825—1834, Bd. i. (1836) S. 227.

(43<) p. 296.— Burkart, Aufenthal, &c. Bd. i. S. 227 and 230.

(a5) p. 296. — Poulett Scrope, Considerations on Volcanos, p. 267; Sir Charles
Lyell, Principles of Geology, 1853, p. 429; Manual of Geology, 1855, p. 580 ;
Daubeny on Volcanos, p. 337. Compare also, on the " elevation hypothesis,"
Dana, Geology, in the Account of the United States Exploring Expedition, vol. x.
p. 369 ; and Constant Prevost, in the Comptes Rendus, t. xli. (1855), p. 866 —
876 and 918-923, sur les e'ruptions, &c. Compare also, on the subject of Jo-
rullo, Carl Pieschel's instructive description of the volcanoes of Mexico, with
elucidations by Dr. Gumprecht, in the Zeitschrift fur Allg. Erdkunde of the
Geographical Society of Berlin, Bd. vi. S. 490 — 517 ; and the recently pub-
lished picturesque views in Pieschel's Atlas der Vulkane der Republik Mexico,
1856, Tab. 13, 14, and 15. The Royal Museum at Berlin possesses in the de-
partment of copperplate engravings and drawings a very fine and numerous
collection of views of Mexican volcanoes (more than forty sheets), taken from
nature by Moritz Rugendas. Of one volcano only, that of Colima, the western-
most of the Mexican volcanoes, this great master has given fifteen coloured views.

(4S6) p. 300. — " Nous avons e'te', M. Bonpland et moi, e'tonne's surtout de
trouver, enchasse's dans les laves basaltiques, lithoides et scorifie'es du Volcan de
Jorullo, des fragmens anguleux blancs ou blancs-verdatres de syenite, compose's
de peu d'amphibole et de beaucoup de feldspath lamelleux. La ou ces masses
ont e'te crevasse'es par la chaleur, le feldspath est deveim filandreux, de sorte que
les bords de la fente sont re'unis dans quelques endroits par des fibres allonge'es
de la masse. Dans les Cordilleres de 1'Ame'rique du Sud, entre Popayan et Al-
maguer, au pied du Cerro Broncoso, j'ai trouve' de ve'ritables fragmens de
gneiss enchasse's dans un trachyte abondant en pyroxene. Ces phe'nomenea
prouvent que les formations trachytiques sont sorties au-dessous de la croute
granitique du globe. Des phe'nomenes analogues presentent les trachytes du
Siebengeblrge sur les bords du Rhin et les couches infe'rieures du phonolithe
(Porphyr-Schiefer") du Biliner-Stein en Boheme." (Humboldt, Essai gdognos-
tique sur le Gisement des Roches, 1823, p. 133 and 339.) Burkart (Aufent-
halt und Reisen in Mexico, Bd. i. S. 230) recognised also, imbedded in the black
oliviniferous lava of Jorullo, " blocks of an altered syenite : hornblende is only
seldom distinctly recognisable. The blocks of syenite would seem to offer irre-

NOTES. CXill

fragable proof of the site of the fiery hearth of the volcano of Jorullo having been
in or below the syenite, which shows itself at the surface for some considerable
extent, a few leagues to the southward, on the left bank of the Rio de las
Balsas which flows into the Pacific." In Lipari, near Caneto, Dolomieu, and,
in 1832, the excellent geognost Friedrich Hoffmann even found in hard masses of
obsidian enclosed fragments of granite formed out of pale-red feldspar, black
mica, and a little light-gray quartz. (Poggendorfs Annalen der Physik, Bel.
xxvi. S. 49.)

(43;) p. 302. — Strabo, lib. xiii. p. 579 and 628 ; Hamilton, Researches in
Asia Minor, vol. ii. chap. 39. The westernmost of the three cones, now called
Kara Devlit, is 530 feet above the plain, and has poured out a great stream of
lava towards Koula. Hamilton counted above thirty small cones in the neigh-
bourhood. The three orifices or abysses (&6&poi and tyvvai of Strabo) are cra-
ters on conical mounts composed of scoria? and lavas.

(438) p. 302. — Erman, Reise urn die Erde, Bd. iii. S. 538 ; Kosmos, Bd. iv.
Anm. 25 (English edition, Note 349). Postels (Voyage autour du Monde, par
le Cap. Lutke, Partie Hist. t. iii. p. 76) and Leopold von Buch (Description
physique des lies Canaries, p. 448) speak of the resemblance to the Hornitos of
Jorullo. Erman, in a manuscript which he kindly communicated to me, de-
scribes a great number of truncated cones of scorise in the enormous lava field
east of the Baidar mountains in the peninsula of Kamtschatka.

(4S9) p. 304. — Porzio, Opera omnia, Med., Phil., et Mathem., in unum col-
lecta, 1736 ; according to Dufre'noy, Me'moires pour servir a une Description
ge'ologique de la France, t. iv. p. 274. In the 9th edition of Sir Charles Lyell's
Principles of Geology, 1853, p. 369, all genetic questions are treated with great
fullness and praiseworthy impartiality. So long ago as 1 749, Bouguer (Figure
tie la Terre, 1749, p. Ixvi.) was not averse from the idea of the volcano of Pi-
chincha having been upheaved : " II n'est pas impossible que le rocher, qui est
biule et noir, ait e'te' souleve' par Faction du feu souterrain." Compare also
p. xci.

(44°) p. 304.— Zeitschrift fur allgemeine Erdkunde, Bd. iv. S. 398.

(441) p. 304. — For the more certain determination of the minerals composing
the Mexican volcanoes, older and more recent collections made by myself and by
Pieschel have been compared.

(442) p. 305. — The fine marble of la Puebla comes from the quarries of
Tecali, Totomehuacan, and Portachuelo, south of the high trachytic mountain
el Pizarro : also near the terrace-pyramid of Cholula, on the road to la Puebla,
I have seen limestone crop out.

.(*4S) p. 306.— The Cofre de P£rote,to the south-east of the Fuerte or Castillo
VOL. IV }L

CX1V NOTES.

de Pe'rote, near the eastern declivity of the great plateau of Mexico, stands al-
most alone, but yet its great mass belongs to an important line of heights which,
forming the margin of the declivity, extends from Cruz Blanca and Rio Frio
towards las Vigas (N. lat. 19° 37' 37"), passing the Cofre de Pe'rote (N. lat.
19° 29', W. long. 97° 08'), west of Xicochimalco and Achilchotla, to the Peak
of Orizaba (N. lat. 19° 2', W. long. 97° 14') in a north and south direction
parallel to the chain (Popocatepetl-Iztaccihuatl) which divides the valley of the
Mexican lakes from the plain of la Puebla. (For the bases of these determina-
tions, see my Recueil d'Observ. Astron. vol. ii. p. 529—532, and 547, as well as
my Analyse de I1 Atlas du Mexique, and Essai Pol. sur la Nouv. Espagne, t. i.
p. 55 — 60.) As the Cofre has been elevated so as to rise abruptly in a pumice-
covered space many miles in breadth, it appeared to me exceedingly interesting
that when I ascended the mountain in the winter, on the 7th Feb. 1804, on the
summit the thermometer sank to 280>4 ; the covering of pumice, of which I
measured the height and thickness barometrically at several points, both in
ascending and descending, was found by me to rise above 780 feet. The lower
limit of the pumice in the plain between Pe'rote and Rio Frio is 7590 feet above
the level of the sea, the upper limit on the northern declivity, 8370 feet ; and
from thence through the Pinahuast, the Alto de los Caxones (12,495 feet high),
where I was able to determine the latitude by the sun's meridian altitude, to the
summit itself, there was no trace of pumice to be seen. When the mountain
was elevated it must have carried with it a portion of the pumice-covering of
the great Arenal, on which it had perhaps been made a smooth flat surface by
water. When on the spot, in February 1804, I made a drawing of this pumice-
covered zone in my journal. It is the same important phenomenon as that de-
scribed in 1 834 by Von Buch at Vesuvius, where horizontal strata of pumice-
tufa have been carried up to a greater height (1900 or 2000 feet), about the
Hermitage del Salvatore. (PoggendorflTs Annalen, Bd. xxxvii. S. 175—179.)
The surface of the dioritic-trachyte rock at the Cofre was not concealed from
observation by snow at the part where I found the pumice highest up on the
mountain. In Mexico, in lat. 19° and 19|°, the limit of perpetual snow is on
an average about 14,760 feet high ; and the summit of the mountain, at the
foot of the small house-like cubical rock where I set up the instruments, is
13,416 feet above the sea. By measurement of angles the cube of rock is 134
feet high, making the absolute summit, which could not be reached on account
of this last vertical precipice, 13,550 feet. I found only some patches of spora-
dically fallen snow, of which the lower limit was 12,150 feet above the sea, fully
700 or 800 feet below the upper limit of forest trees, being fine trees of Pinus
occidentals interspersed with Cupressus sabinoides and Arbutus madrono. The

NOTES. CXV

oak Quereus xalapensis had only accompanied us to 10,338 feet of absolute
height. (Humboldt, Nivellement barometr. des Cordilleres, No. 414—429.)
The name Nauhcampatepetl, which the mountain bears in the Mexican language,
is taken from the same peculiarity of shape which led the Spaniards to give it
the name of Cofre. Nauhcampa, formed from nahui, the number four, no doubt
means four-sided, or quadrangular. Pieschel conjectures the existence of an an-
cient crater-orifice on the eastern declivity of the Cofre of Pe'rote. (Zeitschr.
fur allg. Erdkunde, herausg. von Gumprecht, Bd. v. S. 125.) The view of the
Cofre in my Vues des Cordilleres, PI. XXXIV, was taken by me from near the
Castle San Carlos de Perote, about eight miles distant. The old Aztec name
of Pe'rote was Penahuizapan, signifying, according to Buschmann, a kind of chafer
pinahuiztli (regarded as ill-omened and employed in superstitious augury ; com-
pare Sahagun, Historia gen. de las Cosas de Nueva Espana, t. ii. 1829, p.
10 — 11), a name derived from pinahua, "to be ashamed." From the same
verb comes the above-mentioned name of a district, Pinahuast (pinahuaztli), as
well as the name of a shrub ( ? a Mimosa), pinahuihuiztli, translated by Her-
nandez herba verecunda, the leaves of which droop on being touched.

(444) p. 308. — Strabo, lib. i. p. 58, lib. vi. p. 269, Casaub ; Kosmos, Bd. i.
S. 451 (English edition, Note 225), and Bd. iv. S. 270, and, on the Greek name
for lava, Anm. 82 (English edition, Note 306).

(445) p. 308.— Kosmos, Bd. iv. S. 310 arid Anm. 68 (English edition, Note
392).

(446) p. 308. — La Condamine says : " Je n'ai point connu la matiere de la
lave en Ame'rique, quoique nous ayons, M. Bouguer et moi, campe des semaines
et des mois entiers sur les volcans, et nomme'ment sur ceux de Pichincha, de
Cotopaxi et de Chimborazo. Je n'ai vu sur ces montagnes que des vestiges de
calcination sans lique'faction. Cependant 1'espece de crystal noiratre appele vul-
gairement au Pe'rou Piedra de Gallenaqo (obsidienne), dont j'ai rapporte plu-
sieurs morceaux et dont on voit une lentiUe polie de sept a huit pouces de dia-
metre au Cabinet du Jardin du Roi, n'est autre chose qu'un verre forme par les
volcans. La matiere du torrent de feu qui de'coule continuellement de celui de
Sangai dans la province de Macas, au sud-est de Quito, est sans doute une lave ;
mais nous n'avons vu cette montagne que de loin, et je n'e'tois plus a Quito dans
le temps des dernieres e'ruptions du volcan de Cotopaxi, lorsque sur ses flancs il
s'ouvrit des especes de soupiraux, d'ou 1'on vit sortir a flots des matieres enflam-
me'es et liquides qui devoient etre d'une nature semblable a la lave du Ve'suve."
(La Condamine, Journal de Voyage en Italic, in the Me'moires de 1'Acade'mie
des Sciences, Anne'e 1757, p. 357 ; Histoire, p. 12.) These two examples, es-
pecially the first, are not happily chosen. Sangay was first scientifically exa-

h 2

CXV1 NOTES.

mined by Sebastian Wisse in December 1849. What La Condamine, at a
distance of 1 08 geographical miles, took for glowing lava, and even for an out-
pouring of burning sulphur and petroleum, were glowing rocks and masses of
scoria) following each other closely down the steep descent. (Kosmos, Bd. iv.
S. 303 (English edition, p. 258.) I have not seen on the Cotopaxi, any more
than on Tungarahua, Chimborazo, Pichincha, or on the Purace and Sotara near
Popayan, anything which could be regarded as narrow currents of lava having
flowed from those giant-mountains. The unconnected glowing masses of five or
six feet diameter, often containing obsidian, which have been ejected in eruptions
of Cotopaxi, have been carried by floods of melted snow and ice to far into the
plain, and these form partially radiating diverging lines. Elsewhere, La Con-
damine also says very truly : " Ces eclats de rocher, gros comme une chaumiere
d'lndien, forment des trainees de rayons qui partent du volcan comme d'un centre
commun." (Journal du Voyage a 1'Equateur, p. 1 60.)

(447) p. 309. — Guettard's Memoir on extinct volcanoes was read at the Aca-
demy in 1752, therefore three years before La Condamine's journey to Italy,
but it was not printed until 1756 during that visit. (Sre p. 380.)

(44S) p. 313. — " II y a peu de volcans dans la chaine des Andes (said Von
Buch) qui aient ofiert des courants de laves, et jamais on n'en a vu autour des
volcans de Quito. L'Antisana, sur la chaine orientale des Andes, est le seul
volcan de Quito sur lequel M. de Humboldt ait vu pres dn sommet quelque chose
d'analogue a un courant de laves ; cette coule'e e'tait tout-a-fait semblant a de
Tobsidienne." (Descr. des lies Canaries, 1836, p. 468 and 488.)
(44B) p. 315. — Humboldt, Kleinere Schriften, Bd. i. S. 161.
(4SO) p. 316. — " Nous diffe'rons eritierement sur la pre'tendue coule'e d'Anti-
sana vers Pinantura. Je considere cette coule'e comme un soulevement re'cent
analogue a ceux de Calpi (Yana Urcu), Pisque et Jorullo. Les fragmens tra-
chytiques ont pris une epaisseur plus considerable vers le milieu de la coule'e.
Leur couche est plus epaisse vers Pinantura que sur des points plus rapproche's
d'Antisana. L'etat fragmentaire est un effet du soulevement local, et souvent
dans la Cordillere des.Andes les tremblements de terre peuvent Stre produits
par des tassements." (Lettre de M. Boussingault, en aout , 1 834.) Compare
Kosmos, Bd. iv. S. 219 (English edition, p. 170). In the description of his
ascent of Chimborazo (December 1831), Boussingault says: " The mass of the
mountain consists, in my view, of an enormous heap of fragments of trachyte
piled up on each other without any order. These fragments of a volcano, often
of enormous size, were elevated in a solid state, their edges are sharp and
nothing indicates a state of fusion or even of softening. Nowhere is there ob-
served on any of the equatorial volcanoes anything from which a stream of lava

NOTES. CXY11

could be inferred. Nothing has ever been ejected from these craters but masses
of mud, elastic fluids, and glowing, more or less scorified, blocks of trachyte
which are often thrown to considerable distances." (Humboldt, Kleinere
Schriften, Bd. i. S. 200.) On the first origin of the opinion of the upheaval of
accumulations of solid masses, see Acosta in the Viajes d los Andes ecuatoriales,
por M. Boussingault, 1849, p. 222 and 223. In the opinion of this celebrated
traveller, the movements caused among these fragments by earthquake shocks
and other causes, and the gradual filling up the intervals between them must
occasion a gradual lowering of the heights of the mountains.

(4S1) p. 316.— Humboldt, Asie Centrale, t. ii. p. 296—301. (Gustav Rose,
Mineral-geognostische Eeise nach dem Ural, dem Altai und dem Kasp. Meere,
Bd. i. S. 599.) Long narrow walls or dykes of granite may have been raised
up over fissures in the earliest foldings of the earth's crust, analogous to those
remarkable ones which remain still open, which are found at the foot of the vol-
cano of Pichincha ; the " Guaycos " of the city of Quito, thirty or forty feet
broad. (See my Kl. Schr., Bd. i. S. 24.)

(45-) p. 316. — La Condamine, Mesure des trois premiers degre's du Me'ridien
dans rHe'misphere Austral, 1751, p. 56.

(453) p. 317. — Neither Passuchoa, nor Atacazo, between which the Hacienda
el Tambillo intervenes, enters the limits of perpetual snow. The high margin
of the crater, la Peila, has fallen in on the western side, but shows itself on the
eastern as an amphitheatre. It is said in the country that, near the end of the
sixteenth century, Passuchoa which had been previously active ceased to be so
on the occasion of an eruption of Pichincha, and that the cessation has been
absolute and perpetual : confirming the existence of inter-communication be-
tween the opposite Eastern and Western Cordilleras. The proper basin of Quito,
closed in as by walls, towards the north by a mountain-knot between Cotocachi
and Imbaburo, towards the south by the Altos de Chisinche (between 0° 20' N.
and 0° 40' S. lat.), is for the greater part of its length divided in two by the
ridge of Ichimbio and Poingasi. To the east is the valley of Puembo and
Chillo, and to the west the plain of Inaquito and Turubamba. Proceeding from
north to south we have successively, in the Eastern Cordillera — Imbaburo, the
Faldas de Guamani, and Antisana, Sinchulahua, and the perpendicular black
wall, crowned with tower-like elevations, of Ruminaui (stone-eye) ; and, in the
Western Cordillera — Cotocachi, Casitagua, Pichincha, Atacazo, and Corazon, on
the slopes of which the magnificent Alpine flower, the red Ranunculus Gusmani,
flourishes. It has seemed to me suitable to give here, from personal observa-
tions, a brief description of this important and classic ground for volcanic geo-
logy.

CXV111 NOTES.

(4M) p 319. — It is particularly striking that the great volcano Cotopaxi
which (for the most part indeed only after long intervals of repose) manifests
enormous activity, especially by the devastating inundations which it causes, yet
in the intervals between these outbursts shows no visible smoke or vapour, as
seen either from the Paramo de Pansache or from the high plain of Lactagunga.
Comparison with other equally lofty volcanoes shows that this absence is not to
be explained from the rarity of the air and of the vapour at the great elevation
of 19,000 feet. None of the other Nevados in the equatorial Cordilleras are so
often seen in cloudless grandeur and beauty as the truncated cone of Cotopaxi,
t. e. the portion which rises above perpetual snow. . Its uninterrupted regularity
of form much exceeds that of the Peak of TenerifFe, in which a narrow projecting
wall-like ridge of obsidian runs down from the cone. The upper portion of
Tungurahua may have been once almost as distinguished by regularity of form as
Cotopaxi, but it is now greatly disfigured by the effects of the great earthquake
of the 4th February 1797, called the catastrophe of Eiobamba, by which it was
crevassed ; parts fell in, others glided down, carrying with them the forests which
clothed them, and large accumulations of debris were formed. On Cotopaxi, as
Bouguer had already remarked, there are points at which pumice and snow are
intermingled, and almost form one solid mass. A little unevenness in the snowy
mantle can be discovered on the north-west side, where two cleft-like valleys run
down. One does not see from a distance anything of black rocky ridges running
up towards the summit, although, in the eruptions of the 24th June and 9th De-
cember 1742, a lateral opening showed itself half-way up the snow-covered cone
of cinders. " II s'etoit ouvert une nouvelle bouche vers le milieu de la partie
continuellement neige'e, pendant que la flamme sortoit toujours par le haut du cone
tronque'. (Bouguer, in Figure de la Terre, p. Ixviii. Compare also La Conda-
mine, Journ. du Voy. a 1'Equateur, p. 159.) It is only quite near to the sum-
mit that one perceives some horizontal, parallel, but interrupted black streaks.
Looked at through the telescope under different incidences of light they appear
to me to be ridges of rock. This whole upper part is steeper than the rest of
the mountain, and forms, almost close to the truncation of the cone, a wall-
like annular ridge of unequal height, not, however, visible to the naked eye by
reason of the great distance. My description of this highest, almost perpendi-
cular, encircling ridge strongly drew the attention of two distinguished geolo-
gists, Darwin (Volcanic Islands, 1844, p. 83) and Dana (Geology of the U. S.
Explor. Exped., 1849, p. 356). The volcanoes of the Galapagos islands,
Diana's Peak in St. Helena, and Teneriffe, show analogous conformations. The
highest point of Cotopaxi, of which I measured the elevation by angles, is in a
black convexity. Perhaps it is the inner wall of a higher and more distant

NOTES. CX1X

crater margin ; or is the snowlessness of the projecting rock occasioned both by
steepness and by crater warmth? On one night, in the autumn of 1800, the
whole upper part of the cone appeared luminous, without this being followed by
any eruption or even by any visible emission of vapour. On the other hand, at
the violent eruption, 4 Jan. 1803, when I was staying on the shores of the
Pacific and saw the windows shake in the port of Guayaquil from the vibration
caused by the thundering noise of the volcano 148 geographical miles distant,
the cone entirely lost its snowy covering and wore a threatening aspect. Had
such an effect been observed on any previous occasion ? Very recently we have
learnt from the lady who has wandered round the world with • such admirable
bravery, Madame Ida Pfeiffer (Meine zweite Weltreise, Bd. iii. S. 170), that in
the early part of April 1 854 an outburst took place in which Cotopaxi emitted
thick columns of smoke " through which darted serpentine flames of fire."
May the flames have been the lightnings of a volcanic tempest excited by eva-
poration ? Eruptions have been frequent since 1851.

The great regularity of the snow-covered truncated cone itself renders still
more striking the appearance, at the lower limit of the region of perpetual snow,
where the conical form begins, of a small grotesquely jagged mass of rock, hav-
ing three or four points. Probably by reason of its steepness, the snow lies on
it only here and there in patches. A glance at my drawing (Atlas pittoresque
du Voyage, PL X.) shows these relations more clearly. I approached this dark
gray, probably basaltic, mass of rock most nearly in the Quebrada and Reventa-
zon de Minas. Although for centuries this remarkable-looking mass has borne,
throughout the province, the name of Cabeza del Inga, yet among the coloured
inhabitants (the Indians) two very different hypotheses concerning its origin
prevail ; one simply asserts that this rock is the pointed top of the volcano whicli
has lallen off, without assigning any time for the supposed event, which the
other hypothesis places in the year (1533) in which the Inca Atahuallpa was
strangled; thus connecting it both with the terrible fiery eruptions of Cotopaxi
which ensued in the same year, and have been described by Herrera, and also
with the oracular predictions of Huayna Capac, Atahuallpa's father, on the ap-
proaching fall of the Peruvian empire. It may be asked whether the one thing
common to these two hypotheses, the supposition that the "Cabeza del Inga"
once formed the summit of the cone, may not indicate an obscure remembrance
of a real event, which might be thus preserved, though without any apprehension
of geological ideas. I altogether doubt the correctness of this view. The idea
of a truncated cone having " lost its top," and of this top having been thrown to
a distance unbroken, just as large masses of rock are seen to be hurled afar in
recent eruptions, is one which presents itself not unnaturally to even very un-

CXX NOTES.

educated minds. The terraced pyramid of Cholula, an artificial erection of the
Toltecs, is truncated : the natives, from the impression that it must have been
originally a complete pyramid, have imagined the fable that its top was destroyed
by a stone which fell on it from heaven, and even assert that fragments of this
aerolite were shown to the Spanish Conquistadores. How would it be possible to
place the first eruption of Cotopaxi at a time when the cone of cinders (the
result of a series of eruptions) is admitted, in the story itself, to have been
already existing ? It seems to me probable that the Cabeza del Inga origi-
nated at the spot where it now is : like Yana-Urcu at the foot of Chimborazo,
and— on Cotopaxi itself — the"Morro," south of Sunniguaicu, and north-
west of the little lake Yurak-cocha (white lake in the Qquechhua language).

Kespecting the name of Cotopaxi, I have said, in the first volume of my
Kleinere Schriften (S. 463), that it is only the first part of the word which can
be interpreted in the Qquechhua language, in which " ccotto " means " heap,"
but that " pacsi " is quite unknown. La Condamine (p. 53) interprets the en-
tire name of the mountain, saying : " le nom signifie en langue des Incas masse
brillante" Buschmann, however, remarks that this requires the substitution
for pacsi of pacsa (which is certainly not the same word), meaning brightness,
shining, especially the soft brightness of the moonshine; but, in order to express
the idea of a " shining mass," the genius of the language would further require
the order of the two words to be reversed, " pacsacotto."

(455) p. 319. — Friedrich Hoffmann, in Poggeudorff's Annalen, Bd. xxvi.
1832, S. 48.

(456) p. 320. — Bouguer, Figure de la Terre, p. Ixviii. How often, since the
earthquake of 19th July 1698, has the little town of Lactagunga been destroyed,
and rebuilt out of blocks of pumice-stone from the subterranean quarries of
Zumbalica ! According to the historic documents, consisting of copies df old
ones which had thus perished, and of some more recently saved, in partial rescues,
from the town archives, it appears that such destructions took place in 1703,
1736, 9th Dec. 1742, 30th Nov. 1744, 22nd Feb. 1757, 10th Feb. 1766, and
4th April 1768: seven times in sixty-five years! In 1802, I still found four
fifths of the town in ruins, the result of the great earthquake of Riobamba, 4th
Feb. 1797.

(W7) p. 321. — This diversity was also already recognised by the sagacious
Abich (Ueber Natur und Zusammenhang vulkanischer Bildungen, 1841, S. 83).

(458) p. 321. — The rock of Cotopaxi has essentially the same mineralogical
composition as the volcanoes which are nearest to it, .Antisana and Tungurahua.
It is a trachyte composed of oligoclase and augite. therefore a " Chimborazo-
rock : " a proof of identity of the volcanic rock in the Cordilleras oppo.-ite to

NOTES. CXXI

each other. In the pieces collected by me in 1 802, and by Boussingault in
1831, the mass is partly light or greenish grey, shining like pitchstone, and
translucent at the edges; and partly black, almost like basalt, with great and
small pores with shining sides. The oligoclase sharply defined is embedded
in the mass ; appearing sometimes very distinctly in shining crystals forming
streaks in the planes of cleavage, and sometimes being very minute, and difficult
to recognise. The interspersed augites, which belong essentially to the rock,
are brownish and blackish green, and of very various sizes. Dark laminae of
mica, and black grains of magnetic iron of a metallic lustre, occur, but only
rarely, and probably accidentally. In the pores of a piece in which there is
much oligoclase there is some native sulphur, probably deposited from the all-
pervading sulphurous vapours.

(459) p. 322. — " Le Volcan de Maypo (lat. austr. 34° 15'), qui n'a jamais
rejete' de ponces, est encore e'loigne' de deux journe'es de la colline de Tollo, de
300 pieds de hauteur, et toute compose'e de ponces qui renferment du feldspath
vitreux, des cristaux bruns de mica, et de petits fragmens d'obsiclienne. C'est
done une eruption (inde'pendante), isole'e tout au pied des Andes et pres de la
plaine." (Le'op. de Buch, Description physique des lies Canaries, 1836, p. 470.)

(46°) p. 322. — Federico de Gerolt, Cartas geognosticas de los principales
Distritos minerales de Mexico, 1827, p. 5.

(461) p. 323. — Compare On the Solidification and Formation of the Earth's
Crust, Kosmos, Bd. i. S. 178 — 180, and Anm. 7 on S. 425 (English edition,
p. 161—163, and Note 137. The experiments of Bischof, Charles Deville, and
Delesse have thrown a new light on the folding and crevassing of the earth's
crust. Compare also the ingenious considerations of Babbage on the occasion of
his thermic explanation of the problem presented by the temple of Serapis, on
the north of Puzzuoli, in the Quarterly Journal of the Geological Society of
London, vol. iii. 1847, p. 186; Charles Deville, Sur la Diminution de Densite
dans les Roches en passant de 1'Etat cristallin a 1'Etat vitreux, in the Comptes
Eendus de 1'Acad. des Sciences, t. xx. 1845, p. 1453; Delesse, Sur les Effets de
la Fusion, t. xxv. 1847, p. 545; Louis Frapolli, Sur le Caractere ge'ologique, in
the Bulletin de la Soc. Geol. de France, 2e se'rie, t. iv. 1847, p. 627; and above
all, Elie de Beaumont, in his important work, Notice sur les Systemes de
Montagnes, 1852, t. iii. The following three sections deserve especial attention
from geologists : " Considerations sur les soulevements dus a une diminution
lente et progressive dn volume de la Terre, p. 1330; Sur Vecrasement trans-
versal, nomme refoulement par Saussure, com me une des causes de 1'e'levation
des chaines de montagnes, p. 1317, 1333, and 1346; Sur la contraction que
les roches fondues eprouvent en cristallisant, tendant des le commencement du

CXX11 NOTES.

refroidissement du globe a rendre sa masse interne plus petite que la capacite"
de son enveloppe exterieure, p. 1235.

(tt2) p. 324. — " Les eaux cbaudes de Saragyn, a la hauteur de 5260 pieds,
sont remarquables par le role que joue le gaz acide carbonique qui les traverse a
1'e'poque des tremblements de terre. Le gaz a cette e*poque, comme 1'hydrogene
carbone de la presqu'ile d'Apcherori, augmente de volume, et s'echauffe avant et
pendant les tremblements de terre dans la plaine d'Arddbil. Dans la presqu'ile
d'Apche'ron la temperature s'eleve de 20° jusqu'a 1'inflammation spontane'e au
moment et a 1'endroit d'une Eruption igne'e, pronostique'e toujours par des trem-
blements de terre dans les provinces de Che'makhi et d'Apche'ron." (Abich, in
the Melanges physiques et chimique's, t. ii. 1855, p. 364 and 365.) (Compare
Kosmos, Bd. iv. S. 223; English edition, p. 175.)

(*") p. 324. — W. Hopkins, Researches on Physical Geology, in the Phil.
Trans, for 1839, Ft. II. p. 311 ; for 1840, Pt. I. p. 193; and for 1842, Pt. I.
p. 43 ; and also On the requisite Conditions of Stability of the exterior Surface
of the Earth, in the Theory of Volcanoes in the Report of the Seventeenth Meet-
ing of the British Association, 1847, p. 45—49.

C64) p. 324. — Kosmos, Bd. iv. S. 35—38, Anm. 33—36 (English edi-
tion, Notes 33—36); Naumann, Geognosie, Bd. i. S. 66 — 76; Bischof,
Warmelehre, S. 382; Lyell, Principles of Geology, 1853, p. 536 to 547, and
562. In the very instructive and pleasant work, Souvenirs d'un Naturaliste,
par A. de Quatrefages, 1854, t. ii. p. 464, the upper* limit of the fluid molten
strata is brought up to the small depth of 20 kilometres : " puisque la plupart
des silicates fondent deja a 666° cent." Gustav Rose remarks that " this low
estimate is based on an error. The temperature of 1300° cent., which Mitscher-
lich gives as the melting point of granite (Kosmos, Bd. i. S. 48; English edition,
p. 27, and Note 13), is certainly the least that can be taken. I have several
times had granite put into the hottest parts of the porcelain furnace, and it has
always been but imperfectly fused. Only the mica is fused, with the felspar
forming a glass with many bubbles ; the quartz becomes opaque, but does not
fuse. It is the same with all kinds of rock which contain quartz, and, indeed,
this may be used as a test for discovering quartz in rocks in which it exists in
quantities so small as not to be discernible by the naked eye, ex. gr. in the
Plauen syenite and in the diorite which we brought back with us, in 1829, from
Alapajewsk in the Ural. All rocks which do not contain any quartz, or any
minerals containing as much silica as does granite (for example basalt), when
exposed to the same heat as porcelain, fuse more easily than granite into a per-
fect glass ; but they do not do this over a spirit lamp with a double current of
air, which yet is certainly capable of affording a temperature of 666° cent." In

NOTES. CXX111

Bischofs remarkable experiments in casting a ball of basalt, that substance
appeared, according to some hypothetical assumptions, to require a temperature
of 165° Re'aumur higher than the melting point of copper. (Warmelehre des
Innern unsers Erdkorpers, S. 473.)

(465) p. 326. — Kosmos, Bd. iv. S. 218 (English edition, p. 169). Com-
pare also, respecting the unequal extension and depth of the icy soil, apart from
latitude, the remarkable observations of Franklin, Erman, Kupffer, and, above
all, Middendorff. Kosmos, Bd. iv. S. 43, 47, and 167 (English edition, p.
42—48, and Note 46).

(466) p. 326. — Leibnitz, in the Protogsea, § 4.

(467) p. 327. — On Vivarais and Velay, see the most recent and very exact
researches of Girard, in his " Geologischen Wanderungen," Bd. i. (1856) S. 161,
173, and 214. The ancient volcanoes of Olot were discovered by the American
geologist Maclure in 1808, visited by Lyell in 1830, and well described and re-
presented by him in his Manual of Geology, 1855, p. 535 — 542.

(46S) p. 328. — Sir Kod. Murchison, Siluria, p. 20 and 55-58 (Lyell, Ma-
nual, p. 563).

(469) p. 328. — Scoresby, Account of the Arctic Regions, vol. i. p. 155 — 169,
Tab. V. and VI.

(47°) p. 329. — Le'op. von Buch, Descr. des lies Canaries, p. 357—369; and
Landgrebe, Naturgeschichte der Vulkane, 1855, Bd. i. S. 121 — 136; and on
the encircling ridges of the elevation-craters (caldeiras) in the islands of San
Miguel, Fayal, and Terceira (according to the charts of Vidal), see Kosmos,
Bd. iv. Anm. 84 zu S. 271 (English edition, Note 308). The eruptions of
Fayal (1672) and S. Jorge (1580 and 1808) appear to depend on the prin-
cipal volcano, that of Pico.

(471) p. 329. — Kosmos, Bd. iv. S. 291 (Anm. 27) and 301 (English edition,
257).

(472) p. 329. — Results of Observations on Madeira, by Sir Charles Lyell and
Hartung, in the Manual of Geology, 1855, p. 515 — 525.

(47S) p. 330. — Darwin, Volcanic Islands, 1844, p. 23 ; and Lieut. Lee,
Cruise of the U.S. Brig Dolphin, 1854, p. 80.

(474) p. 330. — See the excellent description of Ascension by Darwin, Volcanic
Islands, p. 40 and 41.

(475) p. 330. — Darwin, p. 84 and 92, on " the great hollow space or valley
southward of the central curved ridge, across which the half of the crater must
once have extended. It is interesting to trace the steps, by which the structure
of a volcanic district becomes obscured, and finally obliterated. (Compare also
Seale, Geognosy of the Island of St. Helena, p. 28.)

CXX1V NOTES.

(476) p. 331. — St. Paul's Rocks. See Darwin, p. 31—33 and 125.

(477) p. 331. — Daussy, Sur 1'existence probable d'un Volcan sous-marin
dans 1'Atlantique, in the Comptes rendus de 1'Acad. des Sciences, t. vi. 1838,
p. 512; Darwin, Volcanic Islands, p. 92; Lee, Cruise of the U.S. Brig Dolphin,
p. 2, 55, and 61.

(47S) p. 332. — Gumprecht, Die vulkanische Thatigkeit auf dem Festlande
von Afrika, in Arabien, und auf den Inseln des rothen Meeres, 1849, S. 18.

(479) p. 333. — Kosmos, Bd. i. S. 456, Anm. 7 (English edition, Note 237).
On the totality of phaenomena, so far as known to us in Africa, see Landgrebe,
Naturgeschichte der Vulkane, Bd. i. S. 195 — 219.

(48°) p. 334. — The height of Demavend above the sea is given by Ainsworth
14,695 feet; but after correcting a barometric result which probably rested on
an accidental clerical error (Asie Centr. t. iii. p. 327), it comes, according to
Oltmann's Tables, to fully 18,634 feet. A still greater height, 20,085 feet, is
given by the doubtless very trustworthy angles of altitude of my friend Captain
Lemm of the Russian Imperial Navy in 1839; but the distance does not rest on
a trigonometric basis, but only on the assumption that the Volcano of Demavend
is 66 versts (104'3 versts = an equatorial degree) from Teheran. It appears,
according to this, that the Persian ever snow-clad Volcano of Demavend, situated
so near the southern shore of the Caspian, and 600 geographical miles from the
Colchian shore of the Black Sea, exceeds the great Ararat by 2984 (or about
3000 feet), and the Caucasian Elboruz by perhaps 1600 feet. On the Volcano
of Demavend, see Ritter, Erdkunde von Asien, Bd. vi. Abth i. S. 551 — 571;
and on the connection of the name Albordj, from the mythical and vague Zend
geography, with the modern Elbourz (Koh Alburz of Kazwini) and Elburuz,
S. 43, 49, 424, 552, and 555.

X481) p. 340.— Asie Centrale, t. ii. p. 9 and 54—58 (Kosmos, Bd. iv. S. 253,
Anm. 61, English edition, Note 285).

(^ p. 340. — Elburuz, Kasbegk, and Ararat, according to communications
from Struve. Asie Centr. t. ii. p. 57. The height given in the text for the
extinct Volcano of Savalan, west of Ardebil, 15,760 English feet, rests on a
measurement by Chanykow. See Abich, in the Me'langes phys. et chim. t. ii.
p. 361. In order to avoid wearisome repetition as to the sources which I have
drawn from, I will state here that all which in the geological section of the
Kosmos relates to the important Caucasian isthmus is taken from writings of
Abich, communicated to me in manuscript by him, 1852 — 1855, with the most
generous and kind permission to make the freest use of them.

(483) p. 340. — Abich, Notice explicative d'une Vue de 1'Ararat, in the Bul-
letin de la Soc. de Ge'ographie de France, 4e seVie, t. i. p. 516.

NOTES. CXXV

(484) p. 349. — Compare Dana's ingenious remarks " On the Curvatures of
Ranges of Islands, whose Convexity, in the Pacific, is almost always turned to
the south or south-east," U.S. Explor. Exped., by Wilkes, vol. x. (Geology by
James Dana) 1849, p. 419.

(485) p. 349. — The island Saghalin, Tschoka, or Tarakai, is called by Japa-
nese sailors Krafto (written Karafuto). It lies opposite to the mouth of the
Amur (the Black River, Saghalian Ula), and is inhabited by a good-humoured,
dark-coloured, and sometimes rather hairy race. Admiral Krusenstern sup-
posed, as did previously the companions of La Pe'rouse (1787), and Broughtun
(1797), that Saghalin is connected with the Asiatic continent by a narrow sandy
isthmus (in 52° 5' N. lat.) ; but we find among the important Japanese infor-
mation obtained by Siebold, by a survey chart laid down in 1808 by Mamia
Rinso, chief of an Imperial Japanese commission, that Krafto is not a peninsula,
but is surrounded by the sea on all sides (Ritter, Erkunde von Asien, Bd. iii.
S. 488). This result of the meritorious Mamia Rinso has been recently (1855)
confirmed by the fact that the Russian fleet, in the Baie de Castries, at Alexan-
drowsk (lat. 51° 29'), therefore south of the supposed isthmus, were able to
retreat into the mouth of the river (lat. 52° 54') just as Siebold had said. In
the strait, where the isthmus was supposed to be, there are at some points only
five fathoms water. The island begins to be politically important on account of
its vicinity to the Amur, or Saghalian River. The name pronounced Karafto w
Krafto is a contraction of Kara-fu-to, i. e. according to Siebold, "the island ad-
jacent to the Kara ; " Kara being a Japano-Chinese designation for the extreme
north of China, Tartary, and " fu," " adjacent to." Tschoka is a corruption of
Tsokai, and Tarakai is taken by a misunderstanding from the name of a single
village, Taraika. According to Klaproth (Asia Polyglotta, p. 301) Taraikai, or
Tarakai, is the native Eino name of the whole island. Compare Leopold
Schrenk's and Captain Bernard Wittingham's remarks in Petermann's Geogr.
Mittheilungen, 1856, S. 176 and 184; and also Perry, Exped. to Japan, vol. i.
p. 468.

(486) p. 350. — Dana, Geology of the Pacific Ocean, p. 16. The coasts of
Cochinchina, from the bay of Tonquin ; those of Malacca, from the bay of Siam ;
and even those of New Holland south of the twenty-fifth parallel also run
nearly north and south.

(487) p. 360. — Compare the translations of Stanislas Julien from the Japanese
Encyclopedia in my Asie Centr. t. ii. p. 551.

C488) p. 360. — Compare Kaart van den Zuid- en Zuidwest-kust van Japan,
door F. von Siebold, 1851.

(489) p.361. — Compare my Fragmens de Ge'ulogieet de Climatologie asiatiques,

CXXV1 NOTES.

t. i. p. 82, which were published immediately after my return from the Siberian
Expedition ; as well as my Asie Centrale, in which I have combated the opinion
expressed by Klaproth, which I had also myself previously entertained, and which
was rendered probable by the connection of the snowy mountains of the Hima-
laya with the Chinese province of Yun-nan and the Nauling, north-west of
Canton. The mountains of Formosa, which are more than 11,700 feet high,
belong, as do the Ta-ju-ling which bound the Foo-kien on the west, to the same
system of north and south fissures as those of Upper Assam in the Birman coun-
try, and the group of the Philippines.

(49°) p. 362. — Dana, Geology of the Explor. Exped. vol. x. p. 540— 545 ; Ernst
Hofmann, Geogn. Beob. auf der Reise von Otto v. Kotzebue, S. 70; Le'op. de
Buch, Descr. Phys. des lies Canaries, p. 435 — 439. Compare Don Antonio
Morati's large and excellent Chart of the Islas Filipinas (Madrid, 1852), in two
sheets.

(491) p. 362. — Marco Polo distinguishes (Part. III. cap. v. and viii.) Giava
minore (Sumatra), where he staid five months, and where he describes the ele-
phants, not found in Java (Humboldt, Examen crit. de 1'Hist. de la Ge'ogr. t. ii.
p. 218), from the earlier-described Giava (maggiore), "la quale, secondo di-
cono i marinai, che bene lo sanno, e 1' isola piu grande che sia al mondo." This
assertion is still true. According to the outlines of the coast of Borneo and
Celebes by James Brooke and Capt. Rodney Mundy, I find the area of Borneo
12,920 (German geographical) square miles, only one tenth of the area of the
continent of New Holland. Marco Polo's accounts of the "much gold and
great, riches which the mercanti di Zaiton e del Mangi bring from thence,"
show that he as well as Martin Behaim on the Nuremberg Globe of 1492, and
Johann Ruysch in his Roman edition of Ptolemy of 1508 (so important for
the history of the discovery of America), means by Java major, Borneo.

C182) p. 363.— Captain Mundy's Chart (Coast of Borneo Proper, 1847) even
gives 14,000 feet. For doubts respecting this elevation, see Junghuhn's Java,
Bd. ii. S. 850. Kina Bailu is not a conical mountain ; its form is much more
like that of those basaltic mountains found in all zones of the earth, which form
a long ridge with two terminal cupolas.

(493) p. 363. — Brooke's Borneo and Celebes, vol. ii. p. 382, 384, and 386.

(494) p. 364. — Homer, in the Verhandelingen van het Bataviaasch Genoot-
schap van Kunsten en Wetenschappen, Deel xviii. (1839) p. 284 ; Asie Centr.
t. iii. p. 534—537.

(495) p. 364.— Junghuhn, Java, Bd. ii. S. 809 (Battalander, Bd. i. S. 39).
C496) p. 364— Kosmos, Bd. iv. Anm. 86 to S. 326 (English edit. Note 410).
C497) p. 365.— Java, Bd. ii. S. 818—828.

NOTES. CXXVll

(498) p. 365.— Java, Bd. ii. S. 840—842.

(499) p. 366.— Java, Bd. ii. S. 853.

(50°) p. 367. — Leop. von Buch, in the Abhandl. der Akad. der Wissensch. zu
Berlin, 1818 and 1819. S. 62 ; Lyell, Princ. of Geology (1853), p. 447, where
a fine representation of the volcano is given.

(MI) p. 368.— Bory de St.-Vincent, Voy. aux quatre lies d'Afrique, t. ii.
p. 429.

(W2) p. 369. — Valentyn, Beschry ving van Oud en Nieuw Oost-Indien, Deel iii.
(1726) p. 70 : Het Eyland St. Paulo. (Compare Lyell, Princ. p. 446.)

(M3) p. 370. — " Nous n'avons pu former," says d'Entrecasteaux, " aucnne
conjecture sur la cause de 1'incendie de 1'Ile d' Amsterdam. L'lle e'toit embrasee
dans toute son etendue, et nous avons bien distinctement reconnu 1'odeur de bois
et de terre brule's. Nous n'avons rien senti qui put faire presumer que 1'em-
brasement flit 1'effet d'un volcan" (t. i. p. 45). " Cependant," he had said be-
fore (p. 43), " Ton a remarque le long de la cote que nous avons suivie, et d'cu
la flamme e'toit assez e'loigne'e, de petites bouffees de fume'e qui sembloient sortir
de la terre comme par jets ; on n'a pu ne'anmoins distinguer la moindre trace de
feu tout autour, quoique nous fussions tres pres de la terre. Ces jets de fumee
se montrant par intervalles ont paru a MM. les naturalistes etre des indices
presqu'assures de feux souterrains." Should we infer that we have here to do
with the ignition of lignites, of which beds covered by basalts and tufa occur so
frequently in the volcanic islands of Bourbon, Kerguebn, and Iceland ? The
name of Surtarbrand employed in Iceland is taken from Scandinavian mythology,
from the Fire- Giant Surtr, who is to set the world on fire. But such earth-fires
do not themselves usually occasion flames. As in modern times the names of
the islands of Amsterdam and of St. Paul's have unfortunately often been con-
founded on maps, it may be well, in order that the phenomena observed in the
one island may not be erroneously ascribed to the other, to remark that originally
(i. e. at the end of the seventeenth century), of the two islands lying nearly
in the same meridian, it was the southernmost which was named St. Paul's and
the northern one Amsterdam. Their discoverer, Vlaming, assigned to the first
the latitude of 38° 40' S., and to the second, 37° 48' S. These data agree re-
markably well in position, as well as in the respective names, with those furnished
a century afterwards by D'Entrecasteaux, in the expedition in search of La Pe-
rouse (Voyage, t. i. p. 43 — 45); namely, for Amsterdam, according to Beau-
temps-Beaupre, 37° 48' N. (78° 13' E.), and for St. Paul, 38° 38'. So close
an agreement must be accidental, as the places of observation were, no doubt,
not exactly the same. On the other hand, Captain Blackwood in his Admiralty
Chart for 1842 gives for St, Paul 38° 44' S. and 77° 39' E. In the maps aj>-

CXXV111 NOTES.

pended to the original edition of Cook's Voyages, ex. gr. of the first and second
expeditions of that circumnavigator of imperishable renown (Voyage to the
South Pole and round the World, Lond. 1777, p. 1), as well as of the third and
last (Voyage to the Pacific Ocean, published by the Admiralty, Lond. 1784, in
2nd ed. 1785), and even in the account of the three expeditions collectively
(A General Chart, exhibiting the Discoveries of Capt. Cook, in this third and
two preceding voyages, by Lieut. Henry Eoberts), the island of St. Paul is
.correctly given as the southern one : nevertheless, in the text of D'Entrecas-
teaux'a Voyage (t. i. p. 44), the complaint is made (whether justly appears to
me more than doubtful after much examination of editions in the libraries of
Paris, Berlin, and Gottingen) that " in the special Chart of Cook's last Expe-
dition Amsterdam Island is placed to the south of St Paul's." Such an inversion
of the names, as intended to be given by the original discoverer Willem de Vla-
ming, is frequently found in maps of the first third cf the present century ; for
example, in the older meritorious Maps of the World of Arrowsmith and Purdy,
1833. This has probably been occasioned rather by : 1. Fault in the Maps of
Cox and Mortimer. 2. The circumstance that in the Atlas to Lord Macartney's
Voyage to China the smoking volcanic island (of which there is a fine view) is
indeed called St. Paul's, and placed in 38° 42' S., but unfortunately it is sub-
joined, " commonly called Amsterdam ; " and, what is still worse, in the Narra-
tive of the Voyage, Staunton and Dr. Gillan constantly use the name of Am-
sterdam for the " island still in a state of inflammation ; " and, after having
given the true latitude in p. 219, even add (p. 226) " that St. Paul is lying to
the northward of Amsterdam." 3. The same confusion as to the names by
Barrow, who in his " Voyage to Cochinchina in the years 1792 and 1793,"
p. 140 — 157, also calls the more southern island, "which gives out smoke and
flames," Amsterdam. Malte-Brun (Pre'cis de la Ge'ographie Universelle, t. v.
1817, p. 146) justly blames Barrow on this account, but he is wrong in also
blaming De Rossel and Beautemps-Beaupre'. Both the latter assign to the
island of Amsterdam, which is the only one of which they give views, the lati-
tude of 37° 47'; and to the island of St. Paul, as being 50' more to the southy
38° 38' (Voy. de D'Entrecasteaux, 1808, t. i. p. 40—46) ; and, to prove that
the drawing in his book represents the true Amsterdam Island of Willem de
Vlaming, Beautemps-Beaupre' gives in his Atlas a copy of the forest-covered
view of Amsterdam from Valentyn. From the circumstance of the celebrated
navigator Abel Tasman having, in 1 642, with Middelburg called the Island of
Tonga-tabu in the Tonga group, in lat, 21|°, Amsterdam (Burney, Chronolo-
gical History of the Voyages and Discoveries in the South Sea or Pacific Ocean,
Pt. III. p. 81 and 437), Tasman has been erroneously cited as the discoverer of

NOTES. CXX1X

Amsterdam and St. Paul in the Indian Ocean. See Leidenfrost, Histor. Hand-
worterbuch, Bd. v. S. 310.

(M4) p. 370.— Sir James Ross, Voyage in the Southern and Antarctic Re-
gions, vol. i. p. 46 and 50 56.

C505) p. 371.— The same, p. 63—82.

(S06) p. 372. — Result of weighings by Professor Rigaud at Oxford, according
to Halley's old proposal. See my Asie Centrale, t. i p. 1 89.

C07) p. 373.— D'Urville, Voy. de la Corvette 1'Astrolabe, 1826 — 1829, Atlas
PI. : I. Polynesie is made to comprise the eastern portion of the Pacific (the
Sandwich Islands, Tahiti, and the Tonga Archipelago ; and also New Zealand).
2. Microne'sie and MeJane'sie take the western part of the Pacific Ocean : Micro-
ne'sie extending from Kauai, the westernmost of the Sandwich Islands, to near
Japan and the Philippines,, and southwards to the equator, comprising the Ma--
rianas or Ladrones, and the Carolinas and Pelew. Isles. 3. Me'lanesie (from
the dark-haired race of men) by its north-west border touching Malaisie, com-
prises the small groups of Viti or Feejee, the New Hebrides, and Solomon
Islands, and the larger islands of New Caledonia, New Britain, New Ireland, and
New Guinea. The names of Oceanie and Polyne'sie, which have been often ap-
plied in so contradictory a manner, were introduced by Malte-Brun in 1813, and
Lesson in 1828.

C508) p. 373.—" The epithet scattered, as applied to the islands of the ocean
(in the arrangement of the groups), conveys a very incorrect idea of their posi-
tions. There is a system in their arrangement as regular as in the mountain
heights of a continent, and ranges of elevations are indicated, as grand and exten-
sive as any continent presents." (Geology, by J. Dana, or United States' Ex-
ploring Expedition under the command of Charles Wilkes, vol. x. (1849) p. 12.)
Dana counts in the whole Pacific, omitting mere rock islets, 350 basaltic or
trachytic, and 290 coral islands. He divides them into twenty-five groups, of
which nineteen have a mean axial direction of N. 50° — 60° W., and six have
one of N. 20°— 30° E. It is an extremely striking circumstance that all this
great number of islands (with a few exceptions as the Sandwich Islands and New
Zealand) lie between the parallels of 23° 28' of north and of south latitude, and
that there remains so enormous a space without islands between the Sandwich
and Nukahiva groups and the American shores of Mexico and Peru. Dana adds
a further consideration, which contrasts greatly with the small number of the
now active volcanoes : namely, that if we may assume the probability that coral
islands, where they are found lying intermediately between islands which are en-
tirely basaltic, rest also on a basaltic foundation, then the number of volcanic
VOL. IV. i

CXXX NOTES.

openings below and above the surface of the sea (sab-marine and sub-aerial) would
be estimated at upwards of a thousand (p. 17 and 24).

C509) p. 374. — Compare, in present volume, p. 247 and Note 359.

(S1°) p. 375. — Dana, Geology of the U. S. Exploring Expedition, p. 208
and 210.

(5U) p. 375. — Dana, p. 193 and 201. The absence of cones of cinders or
ashes is also very remarkable in the Volcanoes of the Eifel which have poured
forth streams of lava. That the summit crater of Mouna Loa can send forth
eruptions of ashes, is proved by the well-assured accounts which the Missionary
Dibble received from the lips of eye-witnesses, and according to which, in the
wars of Kamehameha against the insurgents in 1789, an eruption of hot ashes,
accompanied by an earthquake, spread a darkness like that of night over the
country, (p. 183.) On the volcanic-glass threads (hair of the goddess Pele, who
was supposed, before her settlement in Hawaii, to have inhabited the now ex-
tinct Volcano Hale-a-kala, the " sun-house," in the Island of Maui), see p. 179
and 199—200.

(s12) p. 375. — Dana, p. 205 : " The term Solfatara is wholly misapplied.
A Solfatara is an area with steaming fissures and escaping sulphur vapours, and
•without proper lava ejections; while Kilauea is a vast crater with extensive
lava ejections and no sulphur, except that of the sulphur banks, beyond what
necessarily accompanies, as at Vesuvius, violent volcanic action." The frame
work of Kilauea, the body of the great basin which contains the lava, also by
no means consists of strata of ashes or fragmentary rocks, but of horizontal beds
of lava superposed like limestone. Dana, p. 193. (Compare Strzelecki, Phys.
Descr. of New South Wales, 1845, p. 105—111.)

(513) p. 376. — This remarkable sinking of the surface-level of the lava has
been confirmed by the experience of numerous travellers, from Ellis Stewart and
Douglas to the meritorious Count Strzelecki, Wilkes's Expedition, and the atten-
tively observing Missionary Coan. At the great eruption of June 1840, the
connection of the rising of the lava in the basin of Kilauea with the sudden in-
flammation of the much lower situated crater Arare, was most decided. The
disappearance of the stream of lava poured forth from Arare, its renewed sub-
terranean course and final reappearance in greater mass, do not allow us to
infer its identity with equal certainty, inasmuch as many longitudinal lava-
yielding fissures have opened contemporaneously over the whole declivity of the
mountain below the level of the floor of the Kilauea basin. As respects the in-
ternal constitution of this singular Volcano of Hawaii, it is very noteworthy that,
in June 1832, both the craters, that of the summit and that of Kilauea simul-
taneously, in the one case poured forth, and in the other occasioned, streams of

NOTES. CXXX1

lava, and were, therefore, simultaneously active. (Compare Dana, p. 184, 188,
193, and 196.)

(514) p. 377. — Wilkes, p. 114, 140, and 157; Dana, p. 221. On account
of the perpetual confusion between r and 1, Mouna Loa is often written Mouna
Koa, and Kilauea Kirauea.

(515) p. 377. — Dana, p. 25 and 138.

(516) p. 378. — Dana, Geology of the U. S. Explor. Exped. p. 138. (Com-
pare Darwin, 'Structure of Coral Reefs, p. 60.)

(517) p. 379. — Loop, de Buch, Description phys. des lies Canaries, 1836,
p. 393 and 403—405.

(518) p. 380. — Dana, Geol. of the U. S. Expl. Exp. p. 438—446; and on
the fresh traces of former volcanic activity in New Holland, p. 453 and 457;
and on the numerous basaltic columns in New South Wales and Van Diemen
Island, p. 495—510; and P. E. de Strzelecki, Phys. Descr. of New South
Wales, p. 112.

(519) p. 381. — Ernest Dieffenbach, Travels in New Zealand, 1843, vol. i.
p. 337, 355, and 401. Dieffenbach calls White Island: " a smoking solfatara,
but still in volcanic activity " (p. 358 and 407), and on the map it is stated to
be " in continual ignition."

(52°) p. 381. — Dana, p. 445—448; Dieffenbach, vol. i. p. 331, 339—341,
and 397. On Mount Egmont, see vol. i. p. 131 — 157.

(521) p. 382. — Darwin, Volcanic Islands, p. 125; Dana, p. 140.

(522) p. 382. — L. de Buch, Descr. des lies Canaries, p. 365. We find,
in the above-named three islands, together with plutonic and sedimentary beds,
phonolites and basaltic rocks ; but these may have appeared at the first upheaval
of the island from the bottom of the sea to above the surface of the waters. No
traces either of fiery eruptions within historic times, or of extinct craters, appear
to have been found.

(M3) p. 383. — Dana, p. 343—350.

(524) p. 383. — Dana, p. 312, 318, 320, 323.

(523) p. 383. — L. von Buch, p. 383; Darwin, Volcanic Islands, p. 25; Dar-
win, Coral Reefs, p. 138; Dana, p. 286 — 305, and 364.

(5S6) p. 385. — Dana, p. 137.

(527) p. 585. — Darwin, Vole. Isl. p. 104, 110—112, and 114. If Darwin
says so decidedly that trachytes are entirely wanting in the Galapagos, it is
because he restricts the term trachyte to common felspar, i. e. orthoclase, or to
orthoclase and sanidine (glassy felspar). The curious inbaked or embedded
pieces in the lava of the small and entirely basaltic crater of James Island con-
tain no quartz, although they appear to rest on a plutonic rock. (Compare, in

i-2

CXXX11 NOTES.

present volume, pp. 300 and 331.) Several of the volcanic cones in the Gala-
pagos Islands have, at the mouth, quite as I have seen at Cotopaxi, a narrow
cylindrical parapet. " In some parts, the ridge is surmounted by a wall or
parapet perpendicular on both sides." Darwin, Vole. Isl. p. 83.

(528) p. 386. — L. von Buch, p. 376.

C529) p. 386.— Bunsen, in Lombard's Jahrb. fur Mineralogie, 1851, S. 856;
and in Poggend. Annalen der Physik, Bd. Ixxxiii. S. 223.

("°) p. 387. — Kosmos. Bd. iv. S. 311—313, and Anm. 70 (present volume
.p. 268—269, and Note 394).

(M1) p. 387. — See Pieschel, Ueber die Vulkane von Mexico, in the Zeit-
schrift fur Allg. Erdkunde, Bd. vi. 1856, S. 86 and 489—532. The statement
(S. 86) that " no mortal had ever climbed the steep point of the Pico del Fraile "
is strangely erroneous, and is refuted by Dr. Gumprecht in the same volume (S.
489). The tower-like summit which is only ten feet broad is, indeed, of diffi-
cult attainment; but I gained it, and observed the barometer there on the 29th
of September 1803, and published the observations so long ago as 1807. More-
over, I struck off, and brought home, pieces from the mass of trachyte where it
had been pierced by lightning, glazed on the inside like lightning tubes. Gilbert,
in 1819, in Bd. Ixi. of his Annalen der Physik, S. 261, gave a memoir on these
pieces which had been laid by me before meetings both of the Berlin and Paris
Academies. (Compare also Annales de Chimie et de Physique, t. xix. 1822,
p. 298.) Where the lightning had pierced cylindrical tubes three inches long,
in such a manner that the upper and lower openings could be distinguished
apart, the rock surrounding these openings was also vitrified. I have also, in
my collection, pieces where, as at the Lesser Ararat and Mont Blanc, the whole
surface has been vitrified without tubular perforation. Herr Pieschel, in October
] 852, was the first who ascended the Volcano of Colima with its double sum-
mit, and arrived at the crater from which he then saw hot vapours of sulphu-
retted hydrogen issue in clouds. Sonneschmid, who, in February 1 796, attempted
the ascent of Colima, but without success, reported accounts of a great eruption
of ashes in 1770. In March 1795, glowing scoriae were ejected, presenting at
night the appearance of a pillar of fire. " To the north-west of the Volcano of
Colima, a volcanic branch fissure runs along the shore of the Pacific. Extinct
craters and ancient lava-streams can be recognised in what are called the Vol-
canoes of Ahuacatlan (on the route from Guadalaxara to San Bias) and of
Tepic." (Pieschel, in the volume above referred to, S. 529.)

(532) p. 388. — Kosmos, Bd. iv. S. 392—397 (present volume, p. 348
to 355).

(**) p. 389. — The name of " Grand Ocean " for the basin of the Pacific,

NOTES. CXXX111

introduced by my friend the learned geographer Contre-Amiral de Fleurieu, the
author of the Introduction historique au Voyage de Marchand, has the fault of
giving to a part the name of the whole, and tends, therefore, to cause confusion.
(M4) p. 391. — On the axis of the greatest heights or culminating points,
and of the volcanoes in the tropical zone of Mexico, see Kosinos, Bd. iv. S. 312
and 343 (in the present volume, p. 268 and 299). Compare also Essai pol. sur
la Nouv. Esp. t. i. p. 257—268, t, ii. p. 173; and Ansichten der Natur, Bd. i.
S. 344—350.

(535) p. 392. — By Juan de Onate, 1594. Memoir of a Tour to Northern
Mexico in 1846 and 1847, by Dr. Wislizenus. On the influence which the
peculiar form of the ground (the great magnitude of the table-land) may be ex-
pected to exert on internal traffic and on the intercourse of the tropical zone with
the north, whenever social order, civil liberty, and industry shall be enjoyed,
compare Essai pol. t. iv. p. 38; and Dana, p. 612.

(536) p. 392. — In this tabular view of elevations between Mexico and Santa
Fe', as in the similar, but less complete, one given in rny Ansichten der Natur,
Bd. i. S. 349, the initials B., W., and H. signify the names of the observers:
W. being Dr. Wislizenus, author of a very instructive and scientific work,
Memoir of a Tour to Northern Mexico connected with Colonel Doniphan's Ex-
pedition in 1846 and 1847 (Washington, 1848); B., Oberbergrath Burkart ;
and H., myself. At the time when I was engaged in astronomical determina-
tions of latitude and longitude in the tropical parts of New Spain (from March
1803 to February 1804), and when, from all the materials that I could discover
and discuss, I ventured to prepare a general map of the entire country, of which
map my honoured friend, Thomas Jefferson, then President of the United States,
had a copy made which has since been often much misemployed, there were
as yet no determinations of latitude in the interior of the country, on the
route to Santa Fe', north of Durango (24° 25' N.). According to two manu-
script accounts of travels found by me in the archives of Mexico, of the
engineers Rivera, Lafora, and Mascaro, in 1724 and 1765, containing compass
bearings and estimated partial distances, careful calculation gave for the im-
portant station of Santa Fe 36° 12' N., and 105° 51' W. (from Don Pedro
de Rivera). (See my Atlas ge'ogr. et phys. du Mexique, Tab. VI. : and Essai
pol. t. i. p. 75, 82.) In the analysis of my map, I have carefully stated
this result to be a very uncertain one, inasmuch as, in estimations of dis-
tance and compass bearings, without correction for the magnetic declination,
and with the want of objects in treeless plains without human habitations, on
an extent of 1200 geogr. miles, it cannot be safely assumed that all errors
compensate each other (t. i. p. 127—131). It happens accidentally that

CXXX1V NOTES.

the above result, compared with the latest astronomical determinations, is much
more erroneous in latitude than in longitude, differing in latitude thirty-one
minutes, and in longitude scarcely as much as twenty-three minutes of arc. I
also succeeded in determining, with approximate correctness, by means of com-
binations, the position of the Lake of Timpanogos, now commonly called the
Great Salt Lake; the river which runs into the smaller Utah Lake (a fresh-
water lake) now alone retains the name, being called the Timpanogos River. In
the language of the neighbouring Utah Indians, the river is called Og-wahbe,
and by abbreviation Ogo only; timpan is rock, therefore Timpan-ogo is rocky
river. (Fremont, Expl. Exped. 1845, p. 273.) Buschmann explains the word
timpa as arising out of the Mexican tetl, rock, having discovered in '* pa" a native
North-Mexican substantive ending ; he assigns to ogo the general signification of
water; see his work entitled "Die Spuren der aztekischen Sprache im nordlichen
Mexico, S. 354—356 and 351. The " Great Salt Lake City " of the Mormons
is in 40° 46' N., 112° 04' W. Compare Expedition to the Valley of the
Great Salt Lake of Utah, by Captain Howard Stansbury, 1852, p. 300; and
Humboldt, Ansichten der Natur, Bd. i. S. 346. My map gives " Montagnes de
Sel Gemme " somewhat to the east of the Laguna de Timpanogos, 40° 7' N.,
111° 47' W.; therefore my first conjecture differed in latitude thirty-nine, and
in longitude seventeen minutes. The latest determinations of the position of
Santa Fe', the capital of New Mexico, with which I am acquainted, are: — a.
according to many star altitudes determined by Lieut. Emory (1846), 35° 44'
6"; b. according to Gregg and Dr. Wislizenus (1848), perhaps at a different
spot, 35° 41' 6". Emory's longitude is 7h 4m 18s of time west of Greenwich;
Wislizenus's is 28' of arc less westerly. (New Mexico and California, by Emory,
Dociun. No. 41, p. 36; Wish p. 29.) The error of most maps of the country
about Santa Fd is to make places too northerly. The elevation of the town of
Santa Fe above the sea is, according to Emory, 6422, and according to Wisli-
zenus, fully 6611 Paris feet; mean, 6516 (or 6944 English) feet; therefore
equal to the height of the passes of the Splugen and the St. Gothard in the
Swiss Alps.

(M7) p. 392. — The latitude of Albuquerque is taken from the good special
map entitled Map of the Territory of New Mexico, by Kern, 1851. The eleva-
tion is, according to Emory (p. 166), 4750 feet; but according to AYislizenus
(p. 122), 4859 feet.

(**) p. 392. — On the latitude of Paso del Norte, compare Wiblizenus, p. 125,
Met. Tables VIII. to XII. Aug. 1846.

(**) p. 394. — Compare Fremont, Report of the Exploring Exped. in 1842,
p. 60; Dana, Geology of the U. S. Expl. Exp. p. 611—613; and for South

NOTES. CXXXV

America, Alcide d'Orbigny, Voy. dans 1'Amerique mend. Atlas, PI. VIII. de
Ge'ologie spe'ciale, Fig. 1.

(54°) p. 394. — Respecting this bifurcation, and the correct names of the
eastern and western chains, compare the great Special Map of the Territory of
New Mexico, by Parke and Kern, 1851; Edwin Johnson's Map of Railroads,
1854; John Bartlett's Map of the Boundary Commission, 1854; Explorations
and Surveys from the Mississipi to the Pacific in 1853 and 1854, vol. i. p. 15;
and, above all, the comprehensive and excellent work of Jules Marcou, geologist
of the Southern Pacific R. R. Survey under the command of Lieut. Whipple,
Resume explicatif d'une Carte geologique des Etats Unis et d'un Profil ge'olo-
gique allant de la Valle'e du Mississipi aux Cotes de 1'Oce'an Pacifique, p. 113
— 116; also in the Bulletin de la Societe' Geologique de France, 2e se'rie, t. xii.
p. 813. In the longitudinal valley enclosed by the Sierra Madre and Rocky
Mountains, in 35° — 38°£ N., the single groups, of which the western chain of
the Sierra Madre and eastern chain of the Rocky Mountains (Sierra de Sandia)
consist, bear special names. To the first-named chain belong (proceeding from
south to north) the Sierra de las Grullas, the Sierra de los Mimbres (Wislizenus,
p. 22 and 54), Mount Taylor (35° 15' N.), Sierra de Jemez, and Sierra de San
Juan; in the easternmost chain of the Rocky Mountains may be distinguished
the Moro Peaks, the Sierra de la Sangre de Christo with the eastern Spanish
Peaks (lat. 37° 32'), and the White Mountains which turn north-westward and
enclose the longitudinal valley of Taos and Santa FtS. Professor Julius Prbbel,
whose examination of the volcanoes of Central America I have already spoken of,
has, with much sagacity, developed the indefim'teness of the geographical name
Sierra Madre in the older maps; but, at the same time, in a treatise enticed
Remarks contributing to the physical Geography of the North- American Conti-
nent (Ninth Annual Report of the Smithsonian Institution. 1855, p. 272 — 281)
he has put forward a view to which, after discussing the many now existing
materials, I cannot accede: it is to the effect that the Rocky Mountains are by
no means to be regarded as a continuation of the Mexican highland in the
tropical portion of Anahuac. Uninterrupted mountain-chains, as in the Apen-
nines, the Swiss Jura, the Pyrenees, and a great part of our Alps, are not,
indeed, to be found running from south-south-east to north-north-west between
the nineteenth and the forty-fourth parallels of latitude, from Popocatepetl in
Anahuac to north of Fremont's Peak in the Rocky Mountains ; but the enormous
general elevation of the surface, which increases more and more in breadth
towards the north and north-west, is continuous from tropical Mexico to Oregon ;
and on this swelling high plain, which is the primary and principal geological
phenomenon, single groups of mountains rise over fissures which have taken

CXXXV1 NOTES.

plnce at more recent and very different periods from eacli other, and often in very
different directions. These groups of mountains, thus rising from a high plat-
form, are so connected in the Rocky Mountains as to form a rampart extending
through eight degrees of latitude; and being rendered conspicuous to a great
distance by conical summits, chiefly trachytic, twelve and thirteen thousand feet
high, they produce on the traveller the more impression (it appears to me),
because to the eye they have, illusively, the effect of rising direct from a low-
land plain. Although in the Cordilleras of South America, of which I know a
considerable portion from having myself explored them, it has been customary,
from the time of La Condamine, to speak of a double and triple range (the
Spanish expression " Cordilleras de los Andes " refers to such arrangement and
division); yet it is not to be forgotten that here also the directions of particular
groups, whether as long ridges or as a series of domes, are by no means generally
parallel to each other or to the direction of the entire chain.

(Ml) p. 395. — Fremont, Explor. Exped. p. 281- 288: Pike's Peak, lat.
38° 50', drawn in p. 114; Long's Peak, in 40° 15'; the ascent of Fremont's
Peak. 13,570 feet high, is described in p. 70. The Wind Eiver Mountains take
their name from the sources of a tributary to the Big Horn River, whose waters
unite with those of the Yellow Stone River which falls into the Upper Missouri
in 47° 58' N., 103° 05' W. See the drawings of mountains having much mica-
slate and granite, in pp. 66 and 70. A change of direction takes place in the
chain of the Rocky Mountains between the parallels of Pike's Peak and Lewis'
and Clark's Pass, when it turns more to the west. This circumstance is recalled
for the sake of comparison with the chain of the Ural which, according to the
arduous examinations of my friend and travelling companion Colonel Ernst Hof-
mann, turns eastward near its northern extremity. Its length is fully 1020
geographical miles from the Truchmenian mountain Airuck-Tagh in 48°^ N. to
the Sablja Mountains in 65° N.; in the course of these seventeen degrees of
latitude it deviates but little from the meridian of 59° E. of Greenwich, but in
65° it bends, as aforesaid, to the eastward, so as to reach the meridian of 66° E.
in the parallel of 67°£ N. Compare Ernst Hofmann, Der nbrdliche Ural und
d;is Kusten-Gebirge Pac-Choi, 1856, S. 191 and 297—305, with Humboldt,
Asie Centrale (1843), t. i. p. 447.

(Ms) p. 396.— Kosmos, Bd. iv. S. 321 (present volume, p. 277).

(M3) p. 397. — The Raton Pass, according to an itinerary map of 1855 be-
longing to the General Report of the Secretary of State, Jefferson Davis, is 7180
feet above the sea. Compare also Marcou, Re'sume explicatif d'une Carte ge*ol.
1855, p. 113.

(M4) p. 397. — We distinguish, proceeding from east to west, the ridges of

NOTES. CXXXYll

Zuni, where the Paso de Zuni is still 7944 feet high ; Zuni Viejo, the old mined
Pueblo drawn by Mollhausen in Whipple's Expedition ; and the present inhabited
Pueblo de Zuni. Forty geographical miles north of the last-named place, there
is still a very small isolated volcanic district near Fort Defiance. Between the
village of Zuili and the slope to the little Colorado River (Colorado Chiquito),
there lies uncovered the " petrified forest," of which an excellent description
and drawings have been given by Mollhausen in 1853 in a memoir sent to the
Geographical Society of Berlin. Among the silicified coniferae, fossil tree-ferns
are also interspersed, according to Marcou (Re'sume' explic. d'une Carte geoL
p. 59).

(54:)) p. 398. — All, according to the profiles by Marcou and from the above-
cited itinerary map of 1855.

(546) p. 398. — The French names introduced by Canadian fur-hunters are
in general use in the country and in English maps. The relative positions of
the extinct volcanoes are, according to the latest determinations, as follows : —
Fremont's Peak, 43° 5' N., 110° 8' W.; Trois Tetons, 43° 38' N., 110° 48' W.;
Three Buttes, 43° 20' N., 112° 40' W.; Fort Hall, 43° 0' N., 112° 23' W.

(547) p. 398. — Lieut. Mullan, On the volcanic Formation, in the Reports of
Explor. and Surveys, vol. i. (1855) p. 330 and 348; see also Lambert's and
Tinkham's account of the Three Buttes, in the same, p. 167 and 226 — 230;
and Jules Marcou, p. 115.

(5ls) p. 400. — Dana, p. 616— G21: Blue Mountains, p. 649— 651; Sacra.
mento Butt, p. 630 — 643; Sliasty Mountains, p. 614; Cascade Range. On the
Monte Diablo Range which has broken through volcanic rock, see also John
Trask, On the Geology of the Coast Mountains and the Sierra Nevada, 1854,
p. 13—18.

(Mq) p. 400. — Dana (p. 615 and 640) estimated the volcano St. Helen's at
16.000 feet, and Mount Hood, therefore, at less than that height; according to
others, Mount Hood would have the great elevation of 18,316 feet, 2526 feet
higher than Mont Blanc, and 4746 feet higher than Fremont's Peak in the
Bocky Mountains. Thus, according to the latter statement (Landgrebe, Natur-
geschichte der Vulkane, Bd. i. S. 497), Mount Hood would, on the one hand, be
only 571 feet lower than Cotopaxi; while, according to that of Dana, on the
other hand, it would only be 2550 feet above the highest summit in the Rocky
Mountains. I always think it desirable to call attention to such " variantes
lectiones."

(•*») p. 400. — Dana, Geol. of the U.S. Explor. Exp. p. 640 and 643—645.

(M1) p. 401. — Older various statements of this elevation are: 10,178
feet according to Wilkes, and 13,535 feet according to Simpson.

CXXXV111 NOTES. '

C552) p. 402. — Karsten's Archiv fur Mineralogie, Bd. i. 1829, S. 243.

(553) p. 402. — Humboldt, Essai politique sur la Nouv. Esp. t. i. p. 266;
t. ii. p. 310.

(M4) p. 402. — According to a manuscript which I was enabled to avail
myself of in 1803 in the archives of Mexico, the whole coast from Nootka to
what was subsequently called Cook's Inlet was visited in 1774 by the expedition
of Juan Perez and Estevan Jose' Martinez. (Essai pol. sur la Nouv. Esp. t. ii.
296—298).

(555) p. 406. — In the Antilles, volcanic activity is confined to the lesser
islands; three or four still active volcanoes having broken forth on a rather
curved fissure, approaching to a north and south direction, and tolerably parallel
to the volcanic fissure of Central America. I have elsewhere — on the occasion
of the considerations suggested by the simultaneity of the earthquakes in the
valleys of the Ohio, Mississippi, and Arkansas Rivers with those in the valley of
the Orinoco and on the coast of Venezuela — described the Caribbean Sea as
forming, according to geognostical views, in connection with the Gulf of
Mexico and the great plain of Louisiana between the Alleghanies and the Rocky
Mountains, one great ancient basin. (Voyage aux Regions e'quinoxiales, t. ii.
p. 5 and 19.) This basin is traversed, through its centre, between 18° and
22° N. lat., by a plutonic mountain range from Cape Catoche on the peninsula
of Yucatan to the islands of Tortola and Virgen gorda. Cuba, Haiti, and Porto
Rico form a west and east range running parallel to the granite and gneiss
chain of Caracas. On the other hand, the mostly volcanic small Antilles con-
nect the above alluded to plutonic chain of the greater Antilles and that of the
coast of Venezuela with each other, and close the southern portion of the basiu
on its eastern side. The still active volcanoes of the lesser Antilles are between
the parallels of 13° and 16°£. Proceeding from south to north, they are as
follows : —

The volcano in the island of St. Vincent, estimated sometimes at 3200, some-
times at 5050 feet. After its eruption in 1718, it was at rest until a fresh one
occurred on the 27th of April 1812, sending forth an enormous quantity of lava.
The first movements, near the crater, began in May 1811, three months after
the island of Sabrina had risen from the sea in the Azores. In the mountain
valley of Caracas, 3496 feet above the sea, they began to be faintly felt in
December of the same year. The complete destruction of the large town of
Caracas took place on the 26th of March of the following year, 1812. As the
earthquake which destroyed Cumana on the 14th Dec. 1796 was with reason
ascribed to the eruption of the volcano of Guadeloupe at the end of September
1796, so the destruction of Caracas appears to have been an effect of the reaction

NOTES. CXXX1X

of a more southerly West-Indian volcano, that of the island of St. Vincent. The
dreadful subterranean noise resembling thunder or the sound of a heavy cannonade,
caused by a violent eruption of the last-named volcano on the 30th of April
1812, was heard in the wide grassy plains (the Llanos) of Calabozo and on the
banks of the Rio Apure, 192 geographical miles to the west of its confluence
with the Orinoco. (Humboldt, Voy. t. ii. p. 14.) The volcano of St. Vincent
had not sent forth any lava between 1718 and 1812; on the 30th of April, a
stream of lava flowed from the summit-crater, and in the course of four hours
reached the sea-shore. It is a very curious fact, and one that was confirmed to
me by very intelligent persons engaged in coasting navigation, that the noise was
heard with much greater strength far out at sea than in the immediate vicinity
of the island.

The volcano of the island of Santa Lucia, which is- commonly termed a Sol-
fatara, is only between twelve and eighteen hundred feet high. There are in
the crater several small basins which are periodically filled with boiling water.
In 1766, an eruption of scoriae and ashes is said to have been observed, which is,
indeed, an unusual phenomenon for a Solfutara; for although it appears, from
very well carried out investigations by James Forbes and Poulett Scrope, that
there can scarcely be a doubt as to the fact of an eruption of the Solfatara of
Pozzuoli in 1 198, yet we might be inclined to regard this occurrence as a lateral
effect of the neighbouring principal volcano, Vesuvius. (See Forbes, in the
Edinb. Journal of Science, vol. i. p. 128; and Poulett Scrope, in the Transact,
of the Geol. Soc. 2nd series, vol. ii. p. 346.) Lancerote, Hawaii, and the Sunda
isles present to us analogous examples of eruptions, situated at extreme distances
from the summit-craters which are the proper seats of the volcanic activity. In
the great eruptions of Vesuvius in 1794, 1822, 1850. and 1855, the Solfatara
of Pozzuoli has remained undisturbed (Julius Schmidt, Ueber die Eruption des
Vesuvs im Mai 1855, S. 156); although Strabo (liv. v. p. 245), long before
the breaking out of Vesuvius, speaks, although vaguely, of fire also in the
region of Dicsearchia (Dicsearchia received in Hannibal's time, from the Eomans
who colonised there, the name of Puteoli. Strabo adds, " Some think that,
on account of the bad smell of the water, the whole district, as far as Baiae and
Kymsea, is called so, because it is full of sulphur, fire, and warm water. Some
think that for this reason also Kymaea, Cumanus Ager, was also called Phlegra;"
and afterwards Strabo speaks of fire and water being poured forth, irpoxoas rov

TTVpbs KO.I TOV uSdTOs).

The modern volcanic activity of the island of Martinique in the Montagne
Pelet (4706 feet high according to Dupuget), Vauclin, and the Pitons du Carbet
is still more doubtful. The great outbursts of vapour of January 22, 1792,

Cxi NOTES.

described by Chisholm, and the showers of ashes of August 5, 1851, deserve to
be more closely inquired into.

The Soufriere de la Guadeloupe, according to the older measurements of Amic
and Le Boucher 5435 and 5109 feet high, but according to the most recent and
very exact ones of Charles Sainte-Claire Deville only 4867 feet high, showed
itself to be a pumice-ejecting volcano on the 28th of September 1797, seventy-
eight days therefore before the great earthquake and the destruction of the town
of Cumana. (Rapport fait au Ge'neral Victor Hugues, par Amic et Hapel, sur
le Volcan de la Basse Terre, dans la nuit du 7 au 8 veude'iniaire, an 6, p. 46 ;
Humboldt, Voyage, t. i. p. 316.) The lower part of the mountain is of dioritic
rock; the volcanic cone whose summit is opened is trachyte containing labra-
dorite. The mountain, which, on account of its ordinary condition, is called the
Soufriere, never appears to have sent forth lava in streams, either from the
summit-crater, or from lateral fissures; but the ashes of the eruptions of Sept.
1797, Dec. 1836, and Feb. 1837, examined by the excellent and lamented
Dufre'noy with the accuracy which characterised him, were found to consist of
finely triturated fragments of lava, in which felspathic minerals (labradorite,
rhyakolite, and sanidine), together with pyroxene, were recognised. (See Lher-
minier, Daver, Elie de Beaumont, and Dufre'noy, in the Comptes rendus de
1'Acad. des Sc, t. iv. 1837, p. 294, 651, and 743—749.) Also Deville re-
cognised in the trachytes of the Soufriere small fragments of quartz, together
•with labradorite crystals (Comptes rendus, t. xxxii. p. 675); as Gustav Rose
did hexagonal dodecahedrons of quartz in the trachytes of the Volcano of
Arequipa. (Meyen, Reise urn die Erde, Bd. ii. S. 23.)

The phenomena here described, of the ejection, during a period of short con-
tinuance, of very various mineralogical substances from the fissured apertures of
a Soufriere, remind us forcibly that what we commonly call solfataras, soufrieres,
or fumaroles, are, strictly speaking, only the indications of certain states of vol-
canic activity. Volcanoes which once poured forth lavas, or, failing these, ejected
unconnected scorise of considerable bulk, or lastly, the same scorias reduced by
friction to a state of powder, in a later stage of diminished activity arrive at a
condition in which they furnish only sulphur sublimates, sulphurous acid, and
aqueous vapour or steam. If we were to call them in this state semi- volcanoes,
we might be likely thereby to give occasion to the idea of their being a peculiar
class of volcanoes. Bunsen — to whom, together with Boussingault, Senarmont,
Charles Deville, and Daubre'e, science is indebted for such valuable advances
obtained by the ingenious and happy application of chemistry to geology, and
more especially to volcanic processes — has shown the manner in which, when
in sublimations of sulphur, which almost always accompany volcanic eruptions,

NOTES. Cxli

the sulphurous vapour encounters glowing pyroxenic rocks, sulphurous acid
originates in the partial decomposition of the oxide of iron which those rocks
contain. If thereafter the volcanic activity sinks to lower temperatures, the
chemical activity enters upon a new phase. The combinations of sulphur with
iron, and perhaps with the metallic bases of the earths and alkalies, begin to act
on the aqueous vapour or steam, and, as a result of reciprocal action, there arise
sulphuretted hydrogen, and the products of its decomposition, viz. free hydrogen
and vapours of sulphur. Sulphur fumaroles survive great volcanic eruptions for
centuries. The muriatic acid fumaroles belong to a different and a later period :
it is but rarely that they assume the character of permanent phenomena. The
origination of muriatic acid in crater gases may be inferred to be from the mu-
riate of soda, which so frequently appears as a product of sublimation in vol-
canoes (and in Vesuvius in particular), being decomposed by silicates, at high
temperatures and under the concurrent action of aqueous vapour, into muriatic
acid and soda, which latter combines with the silicates which are present. Mu-
riatic acid fumaroles, which in Italian volcanoes occur not unfrequently on a very
large scale, and which are then usually accompanied by great sublimations of
common salt (muriate of soda), appear to be only very inconsiderable in Iceland.
As the final links in the chronological series of all these phenomena we find lastly
the emanations of carbonic acid gas only. The hydrogen in volcanic gases has
hitherto been almost entirely overlooked. It is present in the vapour spring of
the great solfataras of Krisuvik and Reykjalidh, in Iceland, and at both those
places is combined with sulphuretted hydrogen. As the latter and sulphurous
acid, on coming into contact, mutually decompose each other, setting free the
sulphur, they never can both present themselves at once. It is not rare, how-
ever, to find them in one and the same fumarole field very near each other. If
in the Icelandic solfataras, which have just been named, sulphuretted hydrogen
gas could so little be recognised, on the other hand, it was entirely wanting in
the solfatara state of the crater of Hecla only a short time after the eruption of
1845, therefore in the first phase of volcanic after-action. Neither by smell nor
by reagents could the slightest trace of sulphuretted hydrogen be discovered,
while the abundant sublimation of sulphur made the presence of sulphurous acid
unmistakably recognisable by the smell to a considerable distance. It is true
that on bringing a lighted cigar over the fumaroles those thick clouds of smoke
showed themselves, which Melloni and Piria (Comptes Rendus, t. xi. 1840, p.
352; and PoggendorfFs Annalen, Erganzungsband, 1842, S. 511) have pointed
out as an indication of the smallest traces of sulphuretted hydrogen. As how-
ever it is easy to satisfy one's-self by experiment, that sulphur by itself, when
sublimated with aqueous vapours, also produces the same phenomenon, it remains

Cxlii NOTES.

doubtful whether even a trace of sulphuretted hydrogen accompanied the ema-
nations of Hecla in 1845 and of Vesusius in 1843. (Compare the excellent
and, in geological respects, highly important Memoir of Robert Bunsen, on the
processes of volcanic rock formation in Iceland, in Poggend Ann. Bd. 83, 1851,
S. 241, 244, 246, 248, 250, 254, and 256 : enlarging and correcting the Me-
moirs of 1847, in Wohler's and Liebig's Annalen der Chemie und Pharmacie, Bd.
62, S. 19.) That the emanations of the solfataras of Pozzuoli were not of sul-
phuretted hydrogen, and did not deposit sulphur on contact with the atmosphere,
as had been asserted by Breislak, in his memoir, entitled, Essai Mine'ralogique
bur la Soufriere de Pozzuoli, 1792, p. 128 — 130, had been already remarked by
Gay-Lussac, when we visited the Phlegraean fields together at the time of the
great lava eruption of 1805. The clear-sighted Arcangelo Scacchi (Memorie
Geologiche sulla Campania, 1849, p. 49 — 121) very decidedly denies the ex-
istence of the sulphuretted hydrogen, because Piria's methods of testing appear
to him only to prove the presence of aqueous vapour : " Son di aviso che lo solfo
emane mescolato a i vapori acquei senza essere in chimica combinazione con altre
sostanze." An actual analysis (which I had long wished for and expected) of
the gases emitted by the solfatara of Pozzuoli has been only very recently fur-
nished by Charles Sainte-Claire Deville and Leblanc, and has fully confirmed the
absence of sulphuretted hydrogen. (Cornptes Rendus de 1'Acad. des Sc. t. xliii.
1856, p. 746.) On the other hand, Sartorius von Waltershausen (Physisch-
geographische Skizze von Island, 1847, S. 120) remarked at the eruption cones
of Etna, in 1811, the strong smell of sulphuretted hydrogen, where, in other
years, only sulphurous acid had been perceived. Ch. Deville found a small
portion of sulphuretted hydrogen, not at Girgenti and in the Macalube, but on
the eastern slope of Etna, in the spring of Santa Venerina. It is a striking
circumstance, that, in the important series of chemical analyses which Boussin-
gault made on gas-emitting volcanoes in the chain of the Andes (from Purace
and Tolima to the high plains of Los Pastos and Quito), both muriatic acid and
hydrogene sulfureux are wanting.

(556) p. 406. — Older works give as the number of still burning volcanoes : —
Werner, 193; Caesar von Leonhard, 187; Arago (Astronomic populaire, t. in.
p. 170), 175; all less than iny number, the differences in defect varying
between -J- and ^, and being attributable partly to diversity in the principles
by which the continuance of ignition is judged of, and partly by deficiency in
the data collected. Since, as I have already remarked above, and as historic
experience teaches us, volcanoes, which had been believed to be extinct, have
again manifested their activity after long periods of repose, the numerical result
propounded by me is rather to be regarded as too low than too high. Leopold

NOTES. cxliii

von Buch, in the Appendix to his masterly description of the Canaries, and
Landgrebe. in his Geography of Volcanoes, have not hazarded any general
numerical estimate.

(557) p. 407. — This description is quite opposed to the often repeated repre-
sentation of " Vesuvius according to Strabo," given hi Poggendorff's Annalen der
Physik, Bd. xxxvii. S. 190, Tafel I. A very late writer, Dio Cassius, under
Septimius Severus, was the first to discuss, not the origination of several sum-
mits, as has often been asserted, but the first who tried to show how in course
of time the form of the summit has altered. He recalls (quite in confirmation
of what had been said by Strabo) that formerly the mountain had an everywhere
flat summit. His words (lib. Ixvi. cap. 21, ed. Sturz, vol. iv. 1824, p. 240)
are to this effect: — " For Vesuvius is situated on the sea near Naples, and has
abundant sources of fire. The whole mountain was formerly of equal height,
and the fire rose out of its middle; for in that part only it is burning, its whole
outside being free from fire. Since then the exterior is always without fire,
while the interior is parched with heat and turned into ashes ; the pointed parts
around have hitherto retained their ancient height, while the whole fiery inside,
consumed by time, has sunk down and become hollow, so that, to compare great
things with small, the mountain resembles an amphitheatre." (Compare Sturz,
vol. vi. annot. ii. p. 568.) This is a sufficiently clear description of those parts
of the mountain which, since the year 79, have become the crater-margins. The
interpretation which refers this passage to the Atrio del Cavallo appears to me
incorrect. According to the excellent hypsometric investigation made in 1855
by the active and distinguished Ohntttz astronomer, Julius Schmidt, Punta
Nasone of the Somma is 3772 feet high; the Atrio del Cavallo at the foot of
Puiita Nasone, 2666 feet; and Punta, or Kocca, del Palo (the highest northern
crater-margin of Vesuvius), 3990 feet. My barometric measurements in 1822
(Ansichten der Natur, Bd. ii. S. 290 — 292) gave for the same three points
3746, 2577, and 4022 feet (the differences are 26, 89 £, and 32 feet). The
floor of the Atrio del Cavallo has undergone great alterations of level since the
eruption in February 1850, according to Julius Schmidt (Eruption des Vesuvs
im Mai 1855, S. 95).

(M8) p. 408. — Velleius Paterculus, who died under Tiberius, does, indeed
(ii. 30), name Vesuvius as the mountain which Spartacus occupied with his
gladiators; whereas Plutarch, in the biography of Crassus, cap. ii., speaks only
of a locality amidst rocks having a single narrow entrance. The servile war ut'
Spartacus was in the year of Rome 681, therefore 152 years before Pliny's
eruption of Vesuvius, on the 24th of August 79 A.D. That Florus — a writer
living under Trajan, and who, therefore, being cognizant of the eruption just

Cxl'lV NOTES.

named, was aware of the hidden contents of the recesses of the mountain — «•
should have spoken of it as " cavus," can, as has been already remarked by
others, prove nothing as to its earlier condition. (Florus, lib. i. cap. 16: Vesu-
vius mons, Mtnxi ignis imitator; lib. iii. cap. 20: fauces cavi montis.)

C59) p. 409. — It is in any case certain that Vitruvius wrote earlier than the
elder Pliny; not only because he is cited three times in the list of Pliny's
authorities, wrongly assailed by the English translator Newton (lib. xvi., xxxv.,
and xxxvi.), but because a passage in book xxxv. cap. 14, § 170 — 172, has
been distinctly proved, by Sillig (vol. v. 1851, p. 277) and Brunn (Diss. de
Auctoruin Indicibus Plinianis, Bonuas, 1856, p. 55 — 60), to have been extracted
by Pliny himself from our Vitruvius. Compare also Sillig's edition of Pliny,
vol. v. p. 272. Hirt, in his memoir on the Pantheon, places the writing of
Vitruvius's architecture between the years 16 and 14 B.C.

C860) p. 409. — PoggendorfPs Annalen, Bd. xxxvii. S. 175—180.

(^ p. 409. — Carmine Lippi : " Fu il fuoco o 1' acqua che sotterrb Pompei
ed Ercolano?" 1816, p. 10.

(562) p. 409. — Scacchi, Osservazioni critiche sulla Maniera come fu sepellita
1'antica Pompei, 1843, p. 8—10.

(563) p. 411. — Sir James Ross, Voyage to the Antarctic Regions, vol. i.
p. 217, 220, and 364.

(564) p. 412. — Gay-Lussac, Re'flexions sur les Volcans, in the Annales de
Chiraie et de Physique, t. xxii. 1823, p. 427; Kosmos, Bd. iv. S. 218, English
edition, p. 170; Arago, (Euvres completes, t. iii. p. 47.

(565) p. 413. — Reduced to Timana. the Volcan de la Fragua is in about
1° 48' N., and 75° 30' W. Compare, in the Atlas to my Travels, the Carte
hypsome'trique des Noeuds de Montague dans les Cordilleres, 1831, Pi. V., and
also PI. XXII. and XXIV. This mountain, situated so far to the east, and so
much by itself, deserves to be visited by a geologist who shall also be able to
determine longitudes and latitudes astronomically.

(566) p. 414. — In the three groups which, according to the old geographical
nomenclature, belong to Auvergne, the Vivarais, and the Velay, the distances
stated in the text are taken, in each case, from the northernmost part of the
group to the Mediterranean, between the Golfe d'Aigues Mortes and Cette. In
the first group, that of the Puy de Dome, the northernmost point referred to is
a crater which has broken forth in the granite near Manzat, and is called Le
Gour de Tazena. (Rozet, in the Me'm. de la Sue. Ge'ol. de France, t. i. 1844,
p. 119.) Still more to the south than the group of the Cantal, and therefore
nearest to the sea, at a distance from it of about seventy miles, is the small

NOTES. Cxlv

volcanic district of La Guiolle, near the Monts d'Aubrac, north-west of Chirac.
Compare the Carte ge'ologique de France, 1841.

C87) p. 414. — Humboldt, Asie Centrale, t. ii. p. 7—61, 216, and 335 —
364; Kosmos, Bd. i. S. 254 (English edition, p. 232, 233). I find the Alpine
lake of Issikul, on the northern slope of the Thian-schan, which has only
recently been reached by Russian travellers, already marked on the celebrated
Catalonian Map of 1374, which is preserved as a gem among the manuscripts
of the Paris library. Strahlenberg, in his work entitled " Der nordliche und
ostliche Theil von Europa und Asien" (Stockholm, 1730, S. 327), has the merit
of having been the first to represent the Thian-schan as an independent chain,
but without recognising in it volcanic activity. He gave it the very indefinite
name of Mousart, which, inasmuch as the Bolor chain had the general un-
individualising name of Mustag, signifying snow, gave occasion for another
century to erroneous representations and a bad and contradictory set of names
for the mountain ranges north of the Himalaya, causing the chains which follow
parallels of latitude, and those which follow the direction of the meridian, to be
confounded with each other. Mousart is a corruption of the Tartar word Muz-
tag, which has the same meaning as our Snow mountains, Sierra Nevada of the
Spaniards; Himalaya in the Institutes of Menu, habitation (alaya) of snow
(Mma); and the Sine-schan of the Chinese. Eleven hundred years before
Strahlenberg, under the dynasty of the Sui, in the time of the Frankish king
Dagobert, the Chinese possessed maps, constructed by the order of their govern-
ment, of the countries from the Yellow Eiver to the Caspian Sea, on which the
Kuen-lun and the Thian-schan were drawn. As I think I have shown elsewhere
(Asie Centrale, t. i. p. 118—129, 194—203, and t. ii. p. 413 to 425), it was
these two ranges of mountains, and especially the first, which, when the march
of the Macedonian armies brought the Greeks into nearer acquaintanceship with
the interior of Asia, spread among their geographers the knowledge of a belt of
mountains dividing the continent into two parts, and extending from Asia Minor
to the Eastern Sea, from India and Scythia to Thinje. (Strabo, lib. i. pag. 68,
lib. xi. pag. 490.) Dicsearchus, and after him Eratosthenes, called this chain
the prolongation of the Taurus. The Himalayas were included in this designa-
tion. Strabo says expressly (lib. xv. pag. 689), " India is bounded on the north,
from Ariana to the Eastern Sea, by the extremities of the Taurus, which the
natives call severally Paropamisos, Emodon, Imaon, and other names, but which
the Macedonians call Caucasus." He had said previously (lib. xi. pag. 519), in
the description of Bactriana and Sogdiana: " The last part of the Taurus, which
is called Imaon, touches the Indian (Eastern ?) Sea." The names of " beyond"
and " within the Taurus " referred to the belief in what was supposed to be a
VOL. IV. k

Cxlvi NOTES.

single west and east chain. Strabo says : " The Hellenes term the half of Asia
which slopes to the north the Hither Side of the Taurus, and that towards the
south the Beyond Taurus " (lib. ii. p. 129). But in the later times of Ptolemy,
when commerce, and especially the trade in silk, had become active, the name
Imaus was transferred to a meridian chain, that of Bolor, as is shown by many
passages in the sixth book. (Asie Centr. t. i. p. 146 — 162.) The line in
which, according to Hellenic views, the Taurus mountains parallel to the equator
were supposed to divide the whole continent, was first termed, by Dicasarchus a
' disciple of the Stagirite, " a diaphragm," or dividing-wall, because the geogra-
phical latitude of other points could be measured by perpendicular lines drawn
to it. The Diaphragm was the parallel of Rhodes prolonged, to the westward
to the Pillars of Hercules, and to the eastward to the coast of Thinaa. (Aga-
themeros, in Hudson's Geogr. Gr. Min. vol. ii. p. 4.) The Diaphragm, or
" Divider," of Dicasarchus, alike interesting in geographical and in orographical
respects, passed into the work of Eratosthenes, where he refers to it in the third
book of descriptive geography, in explaining his table of the inhabited world.
Strabo attaches so much importance to this line that (in lib. i. p. 65) he says :
" Its eastern prolongation beyond Thinaa through the Atlantic Sea may possibly
be the site of another inhabited world, or even of several worlds," though he does
not, strictly speaking, predict the existence of such. It may excite surprise that
he should use the word " Atlantic " instead of Eastern Sea, the name usually
given to the Pacific Ocean ; but as our Indian Sea south of Bengal is called in
Strabo " the Atlantic South Sea," the two seas, which were assumed to unite on
the south-east of India, were often confounded. Thus, in lib. ii. p. 130, it is
said, " India, the greatest and most favoured of lands, which terminates at the
Eastern Sea and at the Atlantic South Sea;" and in lib. xv. p. 689, " The
southern and eastern sides of India, which are much greater than the other
sides, run into the Atlantic Sea;" in which last passage, as well as in that
before referred to relative to Thina3 (lib. i. p. 65), the expression of Eastern Sea
seems even to be avoided. Having been constantly occupied, since the year
1792, with the " strike " and " dip" of mountain-strata, and with their relations
to the geographical direction of the mountain-chains, I have thought that I
ought to call attention to the circumstance that the mean parallel of latitude of
the Kuen-lun, in its whole extent, as well as in its western prolongation through
the Hindu Kho mountains, points to the basin of the Mediterranean and the
Straits of Gibraltar (Asie Centr. t. i.p. 118 — 127, and t. ii. p. 115—118); and
that the subsidence of the sea-bed in a great basin which is volcanic principally
at its northern margin, may well be connected with those elevations and foldings
My dear friend of many years, Elie de Beaumont, so profoundly acquainted

NOTES. cxlvii

with all relations of geological direction, is, on grounds of " loxo dromismus,"
opposed to these views. (Notice sur les Syst ernes de Montagnes, 1852, t. ii.
p. 667.)

(56S) p. 415. — Kosmos, Bd. iv. S. 382 (English edition, p. 338).

(569) p. 415. — Compare Arago, "Sur la cause de la depression d'unc
grande partie de 1'Asie et sur le phenomene que les pentes les plus rapides des
chalnes de montagnes sont (ge'neralement) tourne'es vers la mer la plus voisine,"
in his Astronomie populaire, t. iii. p. 1266 — 1274.

(57°) p. 416. — Klaproth, Asia polyglotta, p. 232; and Me'moires relatifs a
1'Asie (taken from the Chinese Encyclopedia published hy the orders of the
Emperor Kanghi in 1711), t. ii. p. 342; Humboldt, Asie Centrale, t. ii. p. 125
and 135—143.

(571) p. 416. — Pallas, Zoographia Eosso-Asiatica, 1811, p. 115.

(372) p. 417. — Volcanic activity, instead of appearing in the Himalaya chain
•which is nearer to the sea (some parts of it, between the colossal Kinchinjinga
and Schamalari, approach within 428 and 376 geogr. miles of the Bay of
Bengal), does not show itself until the third inland parallel chain, the Thian-
schan, almost four times as far; and it has there broken forth among very
peculiar circumstances of neighbouring depressions which have overthrown strata
and caused fissures. We know from the study, by Stanislas Julien, of Chinese
geographical works, that the Kuen-lun, the northern boundary of Thibet, the
Tsi-schi-schan of the Mongols, has, in the hill Schin-khieu, a hollow which
sends forth a constant flame. (Asie Centrale, t. ii. p. 427—467 and 483.)
This phenomenon appears to be quite analogous to that of the Chimera of
Lycia, so often referred to, which has burnt for thousands of years (see above,
Note 375); it is not properly a volcano, but a "fire-spring," which diffuses to a
distance a sweet-smelling odour (derived from naphtha (?)). The Kuen-lun,
which Dr. Thomas Thomson, the learned botanist of Western Thibet (Flora
Indica, 1855, p. 253), treats (just as I have done in my Asie Centrale, t. i.
p. 127, and t. ii. p. 431) as a continuation of the Hindoo-Kho, with which the
Himalaya, coming from the south-east, meets and unites, approaches so near to
the last-mentioned chain at its western extremity, that my friend Adolph
Schlagintweit regards " the Kuen-lun and Himalaya, on the west of the Indus,
not as separate chains, but as forming one mass of mountains." (Report, No. IX.,
of the Magnetic Survey in India, by Adolph Schlagintweit, 1856, p. 61.)
Throughout a distance which extends to 92° E. longitude, towards the " Starry
Sea " or Lake of the Kuen-lun, as we learn by detailed descriptions drawn up
in the seventh century under the dynasty of the Sui (Klaproth, Tableaux hist,
de 1'Asie, p. 204), forms an east and west parallel chain about 7|° of latitude

k 2

cxlviii NOTES.

distant from the Himalaya. The brothers Hermann and Robert Schlagintweit
•were the first who, in the months of July to September 1856, succeeded in the
adventurous attempt of proceeding frcm Ladak across the range of the Kuen-
lun, and reaching the territory of Khotan. According to these always careful
observers, the strike of the highest water-dividing range at the northern boun-
dary of Thibet, and on which the Karakorum Pass, 18,300 feet high, is situated,
is S.E. — N.W., therefore parallel to the portion of the Himalaya, which is over
against it (». e. that to the west of Dhawalagiri). The rivers of Yarkand and
Karakasch, which form part of the great water-system of the Tarim and Lake
Lop, take their rise on the north-eastern declivity of the Karakorum chain.
From the area of this system of waters, the travellers passed over Kissilkorum,
and by the Hot Springs (49° cent., 1100<2 Fahrenheit), to the little Alpine
lake of Kiuk-kiul in the east and west Kuen-lun range. (Report of Mag. Sur-
vey, No. VIII. Agra, 1857, p. 6.)

(57S) p. 418. — Kosmos, Bd. i. S. 27, 48, 181; Bd. iv. S. 34—47, 164—
169, and 369, mit Anm. 39 and 40 (English edition, vol. i. p. 27, 46, 131;
present volume, Notes 33 — 47, 463, and 464).

(574) p. 418. — Arago (Astron. populaire, t. iii. p. 248) assumes almost the
same thickness for the crust of the earth : 40,000 metres, about 5^ German
geogr. miles, or 22 English; Elie de Beaumont (Systemes de Montagnes, t. iii.
p. 1237) makes the thickness one fourth greater. • The earliest-made assumption
was that of Cordier, in mean value 14 German or 56 English geographical miles:
Hopkins's Mathematical Theory of Stability would require it to fall between
688 and 860 English geographical miles. On geological grounds, I fully con-
cur with Naumann's doubts of so enormous a distance between the fluid in-
terior and the craters of active volcanoes. See his excellent Lehrbuch der
Geognosie, Bd. i. S. 62— -64, 73—76, and 289.

(575) p. 419. — A very remarkable example of the manner in which appre-
ciable alterations of composition may take place, through the gradual accumu-
lation of very minute quantities, has been recently presented by Malagute's
discovery, confirmed by Field, of the presence of silver in sea-water. Notwith-
standing the enormous extent of the ocean, and the small surface of the hulls of
the ships which traverse it, the trace of the silver in the sea-water has become
perceptible on the copper-sheathing of vessels.

(576) p. 420. — Bunsen, Ueber die chemischen Prozesse der vulkanischeu
Gesteinsbildungen, Poggend. Annalen, Bd. Ixxxiii. S. 242 and 246.

(577) p. 420. — Comptes rendus de 1'Acad. des Sciences, t. xliii. 1856, p. 366
arid 689. The first exact analysis of the gas which issues forth with noise
from the great Solfatara of Pozzuoli, and which was collected with much difil-

NOTES. Cxlix

culty by Charles Sainte-Claire Deville, gave : of sulphurous acid (acide sulfu-
reux), 24-5; oxygen, 14'5; and nitrogen, 61 '4.

(578) p. 420. — Kosmos, Bd. iv. S. 255—261. (English edition, present
volume, p. 209 — 215).

(579) p. 420. — Boussingault, Economic rurale (1851), t. ii. p. 724,726:
" La permanence des orages dans le sein de 1'atmosphere (sous les tropiques) est
un fait capital, parce qu'il se rattache a une des questions les plus importantes
de la physique du globe, celle de la fixation de 1'azote de 1'air dans les etres
organise's. Toutes les fois qu'une serie d'etincelles electriques passe dans 1'air
humide, il y a production et combinaison d'acide nitrique et d'ammoniaque. Le
nitrate d'ammoniaque accompagne constammeut 1'eau des pluies d'orage, et
comme fixe par sa nature il ne saurait se maintenir a 1'e'tat de vapeur; on
signale dans 1'air du carbonate ammoniacal, et 1'ammoniaque du nitrate est
amene'e sur la terre par la pluie. Ainsi, en definitive, ce serait une action
electrique, la foudre, qui disposerait le gaz azote de I'atmosphere a s'assimiler
aux etres organise's. Dans la zone equinoxiale, pendant 1'anne'e entiere, tous
les jours, probablement me"me tous les instants, il se fait dans 1'air une continuite
de de'charges e'lectriques. Un observateur place a 1'e'quateur, s'il e'tait doue'
d'organes assez sensibles, y entendrait continuellement le bruit du tonnerre."
But sal-ammoniac, as well as muriate of soda, is occasionally found, as a subli-
mation product of volcanoes, on streams of lava themselves : on Hecla, Vesuvius,
and Etna; in the Guatemala volcanic chain (on the volcano of Izalco) ; and,
above all, in Asia in the volcanic chain of the Thian-schan. The inhabitants of
the district between Kutsch, Turfan, and Hami pay their tribute to the emperor
of China in some years in sal-ammoniac (in Chinese nao-scha, in Persian nu-
schaden), which is an important article of foreign trade. (Asie Centrale, t. ii.
p. 33, 38, 45, and 428.)

(5SO) p. 421. — Viajes de Boussingault (1849), p. 78.

(581) p. 421. — Kosmos, Bd. i. S. 295 and 469 (English edition, p. 270, and
Note 326).

(ffi2) p. 422. — Eozet, Me'moire sur les Volcans d'Auvergne, in the Me'moires
de la Soc. Ge'ol. de France, 2" se'rie, t. i. 1844, p. 64 and 120—130 : " Les
basaltes (comme les trachytes) ont perce' le gneiss, le granite, le terrain houiller,
le terrain tertiaire, et les plus anciens depots diluviens. On voit meme les
basaltes recouvrir souvent des masses de caillous roule's basaltiques ; ils sent
sortis par une infinite d'ouvertures dont plusieurs sont encore parfaitement (?)
reconnaissables. Beaucoup pre'sentent des cones de scories plus ou moins con-
side'rables, mais on n'y trouve jamais des crateres semblables a ceux qui ont
donne des coule'es de laves. . . . ."

Cl NOTES.

(^ p. 422. — Like the granitic pieces enveloped in the trachytes of Jorullo.
Kosmos, Bd. iv. S. 345 (English edition, present volume, p. 300 — 301).

(5S4) p. 422. — Also in the Eifel, according to the high authority of Berg-
hauptmann von Dechen. Kosmos, Bd. iv. S. 281 (present volume, p. 235).

(58S) p. 422. — Kosmos, Bd. iv. S. 357 (present volume, p. 312). The Rio
de Guaillabamba flows into the Kio de las Esmeraldas. The village of Guailla-
bamba, near which I found the isolated basalts containing olivine, is only 6908
feet above the sea. The heat in this valley is extremely oppressive, and it is
still worse in the Valle de Chota, between Tusa and the Villa de Ibarra, of
which the lowest part is only 5288 feet above the sea, and which may rather be
called a deep cleft than a valley, being rather under 9600 feet wide, and fully
4800 feet deep. (Humboldt, Kec. d'Observ. astronomiques, vol. i. p. 307.) The
eruption of fragments called Volcan de Ansango, on the declivity of Antisana,
is not a formation of basalt; they consist of a trachyte containing oligoclase, and
bearing some resemblance to a basalt. On the mutual avoidance, or " antago-
nisme des basaltes et des trachytes," see my Essai ge'ognostique sur le gisement
des Eoches, 1823, p. 348 and 359, and, in general, p. 327—336.

(588) p. 425. — Sebastien Wisse, Exploration du Volcau de Sangay, in the
Comptes rendns de 1'Acad. des Sciences, t. xxxvi. (1853) p. 721 ; compare also
Kosmos, Bd. iv. S. 292, Anm. 40, and S. 301 — 303 (English edition, p. 248,
and Note 364). According to Boussingault, the erupted fragments brought back
by Wisse, and which were collected on the upper portion of the declivity of the
cone (the traveller reached a height of 960 feet below the summit, which has
itself a diameter of 486 feet), consist of a black pitch-like substance, in which
are embedded crystals of glassy (?) felspar. A very remarkable phenomenon,
unique, I believe, in volcanic eruptions, so far as we are at present aware, is that,
together with these large black pieces of trachyte, small angular pieces of pure
quartz are erupted. These fragments (according to a letter written, in January
1851, by my friend Boussingault) are not more than four cubic centimetres in
volume. In the trachytic masses themselves there is no quartz interspersed.
All volcanic trachytes which I have examined in the Cordilleras of South
America and Mexico, and even those trachytic porphyries in which the rich
silver-veins of Real del Monte, Moran, and Regla, north of the high valley of
Mexico, occur, are entirely without quartz. Notwithstanding this apparent
antagonism between quartz and trachyte in burning volcanoes, I am not in-
clined to deny the volcanic origin of the millstone trachytes (trachytes et
porphyres meulieres) to which Beudant has justly called attention. The manner
in which these have burst forth from fissures is, however, doubtless a mode of

NOTES. cli

origin altogether different from the formation of conical or dome-shaped trachytic
frameworks.

(587) p. 425.— Kosmos, Bd. iv. S. 276—280 (English edition, p. 230—234).

(5S8) p. 426. — The most complete account, based on actual measurements
of heights and angles of inclination and on sections, which we possess for
any volcanic district, is that which we owe to the fine investigations of the
Olmiitz astronomer Julius Schmidt, of Vesuvius, the Solfatara, the Monte'
Nuovo, the Astroui, Eocca Monfina, and the ancient volcanoes of the Papal
States (in the Alban Hills, Lago Bracciano, and Lago di Bolsena); see Julius
Schmidt's work, " Die Eruption des Vesuvs im Mai 1855 ;?) and in the accom-
panying Atlas, Plates III., IV., and IX.

(589) p. 426. — In the progressive advances which have been made in our
knowledge of the surface of the moon, from Tobias Mayer to Lohrmann, Madler,
and Julius Schmidt, the belief in the great analogies between the terrestrial and
lunar volcanic frameworks has, on the whole, rather lessened than increased:
not so much on account of relations of dimension and early recognised inter-
arrangement of so many annular forms, as on account of the " rills " and of the
" systems of rays," which cast no shadows, and are more than four hundred miles
long, and from two to sixteen miles broad: as in Tycho, Copernicus, Kepler, and
Aristarchus. It is interesting to notice that Galileo, in his letter to Pater
Christopher Grienberger " Sulle Montuosita della Luna," compared the lunar
King Mountains, whose diameters he believed to be greater than they really are,
to the mountain-encircled land of Bohemia; and that the ingenious Eobert
Hooke, in his Micrographia, ascribed the circular type, which prevails so largely
on the moon's surface, to the reaction of the interior of the lunar globe upon its
exterior. (Kosmos, Bd. ii. S. 508; and Bd. iii. S. 508 and 544: English edition,
vol. ii. p. 349; and vol. iii. p. 365, and Note 588.) In regard to the Eing Moun-
tains of the moon, I have, in recent years, felt a lively interest in the question
of the relative heights of the central mounts and the encircling ridges or crater-
margins, and in the existence of parasitic craters on the encircling ridge itself.
The result of all the careful observations of Julius Schmidt, who is engaged in
continuing and completing Lohrmann's Topography of the Moon, is to the effect:
— " That in no single case does the one central mount attain a height equal to
that of its surrounding crater- wall; and that it is probable that in all cases the
summit of the central mount is even considerably below the surface of the moon
from which the crater has broken forth." Whereas the cone of scoriaa within
the crater of Vesuvius, which rose up on the 22nd of October 1822. according
to Brioschi's trigonometrical measurement passes the Punta del Palo (the high-

Clii NOTES.

est part of the crater-margin) by about thirty feet, and can be seen from Naples,
on the moon, on the contrary, many of the central mounts, as measured by
Madler and by Schmidt, are fully 1000 toises (about 6400 feet) lower than the
mean height of the encircling ridge, and even 100 toises below what may
be supposed to be the mean level of the general surface of that part of the
moon. (Madler, in Schumacher's Jahrbuch for 1841, S. 272 and 274; and
Julius Schmidt's " Der Mond," 1856, S. 62.) The lunar central mounts, or
it may perhaps rather be said " central masses of mountains," have usually
several summits: as in Theophilus, Petavius, and Bulliald. There are six
central mounts in Copernicus, while Alphons alone shows one regular central
peak with a sharp-pointed summit. The above relations remind us of the
Astroni in the Phlegrsean Fields, to whose dome-shaped central masses Von
Buch justly attributed great importance. " These masses (like the central
masses in the Ring Mountains in the moon) did not burst forth; no permanent
connection with the interior, no volcano was formed; but rather, as it were, a
model on a small scale of those great trachytic unopened domes, as Puy de Dome
and Chimborazo, which are so variously distributed over the earth's surface."
(PoggendorfFs Annalen, Bd. xxxvii. 1836, S. 183.) The encircling margin of
the Astroni has everywhere the form of a closed ellipse, never rising higher than
130 toises (831 feet) above the level of the sea. The summits of the central
domes are 658 feet lower than the highest part of the south-western crater-wall.
The domes form two parallel ridges clothed with thick bushes. (Julius Schmidt,
" Eruption des Vesuvs," S. 147; and the same author's " Der Mond," S. 70 and
103.) One of the most remarkable objects on the moon's surface is the Ring
Mountain Petavius, in which the whole interior crater-floor has expanded in a
convex form, and yet is crowned by a central~mount. The convexity, resembling
a blister-like swelling, is here a permanent form. In our terrestrial volcanoes
it is only temporarily that the crater-floor is sometimes so inflated by the force
of the vapours beneath as to rise almost to the height of the crater-margin ; but
as the vapours break through and burst forth, the inflated floor sinks down
again. The greatest diameters of terrestrial craters are those of the Caldeira de
Fogo, according to Charles Deville 4100 toises (4'3 geogr. miles); and the Cal-
deira of Palma, according to Leopold von Buch, 3100 toises; whereas, on the
moon, Theophilus is 50,000 toises, and Tycho 45,000 toises (or respectively 52
and 45'2 geogr. miles). Parasitical secondary craters, which have broken forth
on a marginal wall of a great crater, are very frequent on the moon. The crater-
floor in these parasites is usually empty, as on the great rent-asunder margin
of Maurolycus; it is more rare to see a small central mount, perhaps a cone of
eruption, as in Longomontanus. In a fine sketch of the crater-system of Etna,

NOTES. cliii

sent to me, in August 1854, from Flensburg, by my friend the astronomer
Christian Peters (now at Albany in North America), one can distinctly recognise
the parasitical margin-crater (called Pozzo di Fuoco) which was formed on the
E.S.E. side in January 1833, and had several strong eruptions of lava until
1843.

(59°) p. 427. — The little-characteristic indefinite name of trachyte (rough
stone), which is now so generally given to the rock in which volcanoes break
forth, was first applied by Hauy in 1822, in the second edition of his Traite' de
Mineralogie, vol. iv. p. 579, to a rock of Auvergne, and merely accompanied by
a notice of the derivation of the name, and a short description in which there is
no mention of the older designations : " granite chauffe' en place " of Desmarets,
"trap-porphyry," and "domite." Previously to 1822, however, the name had
been known in verbal communications occasioned by Hauy's lectures in the
Jardin des Plantes ; and it appears in Von Buch's Memoir on Basaltic Islands
and Elevation-Craters, published in 1818; in Daubuisson's Traite' de Mine'ralogie,
1819; and in Beudant's important work, " Voyage en Hongrie." From friendly
letters very recently received from Elie de Beaumont, the recollections of M.
Delafosse, formerly Aide-Naturaliste to Hauy, and now Membre de 1'Institut,
would carry the use of the name back to between 1813 and 1816. The publi-
cation of the name " domite," by Von Buch, seems, according to Ewald, to belong
to 1809; domite is first mentioned in the third letter to Karsten (Geognostische
Beobachtungen auf Reisen durch Deutschland und Italien, Bd. ii. 1809, S. 244).
It is there said : — " The porphyry of the Puy de Dome is a peculiar kind of
rock, hitherto without a name, consisting of felspar crystals with glassy lustre,
hornblende, and black lamina} of mica. In the cracks and clefts of this rock,
which I will call provisionally domite, there are fine drusic cavities whose walls
are covered with crystals of ferruginous mica. In the whole length of the Puy,
cones of domite alternate with cones of scoria;." The second volume of the
Travels, which contains the letters from Auvergne, was printed in 1806, but not
published until 1809, which is therefore, strictly speaking, the date of the
publication of the name. It is singular that in Von Buch's memoir four years
later, " Ueber den Trapp-Porphyr," domite is no longer spoken of. In referring,
in the text, to a profile of the Cordilleras contained in the journal of my travels
for July 1802, extending from 4° N. to 4° S., and inscribed " Affinite' entre le
Feu volcanique et les Porphyres," my wish was to recall that it was this profile
— which represents the three breakings-through of the volcanic groups of
Popayan, Los Pastes, and Quito, and the outburst of trap-porphyry in the granite
and mica-slate of the Paramo de Assuay (near the Cadlud Road, at a height of
15,526 feet), — which led Leopold von Buch to ascribe to me only too decidedly

cllV NOTES.

and kindly the first recognition that " all the volcanoes of the Andes have their
seat in a porphyry which is a peculiar rock, and belongs essentially to volcanic
formations." (Abhandlungen der Akademie der Wiss. zu Berlin, for 1812 —
1813, S. 131, 151, and 153.) I may, indeed, have expressed the phenomenon
with most generality; hut, as early as 1789, Nose, whose merits were long un-
recognised, had, in his " Orographic Letters," described the volcanic rock of the
Siebengebirge as " a peculiar kind of Rhenish porphyry, nearly allied to basalt and
porphyritic schist." He says that this formation is particularly characterised
by glassy felspar, which he proposes to call sanidine, and that, by its age, it
belongs to the middle-fketz strata. (Niederrheinische Reise, Th. I. S. 26, 28,
and 47; Th. II. S. 428.) I am inclined to think that Von Buch was mistaken
in stating that Nose recognised this porphyritic formation (which he not very
happily calls " granite-porphyry ") and basalts as being younger than the most
recent floetz-strata. The great early-removed geognost said: " The whole rock
ought to be named from the glassy felspar (therefore sanidine-porphyry), if it
had not already the name of trap-porphyry." (Abh. der Berl. Akad. for 1812
— 1813, S. 134.) The history of the systematic nomenclature of a science is
not without some importance, inasmuch as it reflects the succession of prevailing
opinions.

(WI) p. 427. — Humboldt, Kleinere Schriften, Bd. i. Vorrede, S. iii.— v.

(M2) p. 428. — Leop. von Buch, in Poggendorff's Annalen, Bd. xxxvii. 1836,
S. 188 and 190.

C88) p. 428. — Gustav Rose in Gilbert's Annalen, Bd. 73, 1823, S. 173, and
Annales de Chimie et de Physique, t. xxiv. 1823, p. 16. Oligoclase was first
proposed as a new mineral species by Breithaupt. (Poggendorff's Annalen, Bd.
viii. 1826, S. 238.) Subsequently oligoclase appeared to be identical with
a mineral which Berzelius observed in a granite dyke in gneiss, near Stock-
holm, and which, on account of similarity of chemical composition, he had called
" Natron Spodumen." (Poggendorff's Ann. Bd. ix. 1827, S. 281.)

(594) p. 429. — See Gustav Rose on the granite of the Riesengebirge, in Pogg.
Ann. Bd. Ivi. 1842, S. 617. Berzelius had only found oligoclase, his " Natron
Spodumen," in a granite dyke ; it is in the above-cited memoir that it is first
spoken of as an ingredient of granite (of the rock itself). Gustav Rose deter-
mined oligoclase by its specific weight, the larger quantity of lime contained in
it in comparison with albite, and its greater fusibility. A specimen, of which
he had found the specific gravity 2*682, was analysed by Rammelsberg. (Iland-
wbrterbuch der Mineral. Suppl. i. S.I 04 ; and G. Rose, Ueber die zur Granitgruppe
gehorenden Gebirgsarten, in der Zeitschr. der Deutschen geol. Gesellschaft, Bd.
i. 1849, S. 364.)

NOTES. civ

(595) p. 430. — Rozet, sur les Volcans de 1'Auvergne, in the Mem. de la Soc.
Ge'ologique de France, 2eme Se'rie, t. i. p. i. 1844, p. 69.

(596) p. 430. — Fragments of leucite-ophyr, collected by me on Monte Nuovo,
have been described by Gustav Rose in Fried. Hoffmann's Geognostischen Beo-
bachtungen, 1839, S. 219. On the trachytes of the Monte di Procida, in the
island of the same name, and on the rocky shoal of S. Martino, see Roth, Mono-
graphic des Vesuvs, 1857, S. 519 — 522, Tab. viii. The trachyte of the island
of Ischia contains, in the Arso or lava-current of Cremate (1301), glassy felspar,
brown mica, green augite, magnetic iron, and olivine (S. 528) ; no leucite.

(597) p. 430. — The geologo-topographical relations of the Siebengebirge
near Bonn have been developed and described, with generalising sagacity and
great exactness, by my friend Berghauptmann H. von Dechen, in the 9th
year's series of the Verhandlungen des natur-historischen Vereines der preus.
Rheinlande und Westphalens, 1852, S. 289—567. All the chemical analyses
of the trachytes of the Siebengebirge which have yet appeared are collected there
(S. 323 — 356): and some account is given of the trachytes of the Drachenfels
and Rb'ttchen, in which, besides the large crystals of sanidine, many small crys-
talline particles can be distinguished in the ground mass. " These were recog-
nised by Dr. Bothe, by chemical analyses performed in Mitscherlich's laboratory,
to be oligoclase ; in entire agreement with the oligoclase of Danvikszoll (near
Stockholm), cited by Berzelius." (Dechen, S. 340 — 346.) The Wolkenberg
and the Stenzelberg are without glassy felspar (357 and 363), and belong not
to the second division but to the third; they present a Toluca rock. Many new
views are contained in the section of the " Geognostischen Beschreibung des Sie-
bcngebirges," which treats of the relative age of trachytic and basaltic conglome-
rates (S. 405 — 461). " Besides the more rare dykes of trachyte, in the trachytic
conglomerates, which show that after the deposition of the conglomerate the for-
mation of trachyte still continued to take place (S. 413), there are frequent
dykes of basalt (416). The formation of basalt decidedly comes down to a later
period than that of trachyte, and the principal mass of basalt is here younger
than the trachyte. On the other hand, it is only a portion of this basalt, — not
all basalt (S. 323), — which is younger than the great mass of the brown coal.
The two formations, basalt and brown coal, run into each other in the Siebenge-
birge, as in so many other places, and are to be looked upon, on the whole, as co-
temporaneous." Where very small crystals of quartz appear as a rarity in the
trachytes of the Siebengebirge, as they do according to Noggerath and Bischof
in the Drachenfels and in the Rhondorfer Thai, they fill cavities, and seem to be
of later formation (S. 361 and 370) : perhaps they have been produced by the
weathering of the sanidine. On Chimborazo I once saw similar but very thin

Clvi NOTES.

depositions of quartz on the sides of the cavities in some very porous brick-red
trachytic masses, at an elevation of about 17,000 feet. (Humboldt, Gisement
des Roches, 1823, p. 336.) These pieces, which are repeatedly mentioned in
my journal, are not in the Berlin collections. Weathering of oligoclase, or of the
•whole ground-mass of the rock, may also give rise lo such traces of free silicic
acid. Some points in the Siebengebirge still deserve fresh and persevering ex-
amination. The highest summit, the Lbwenberg, spoken of as basalt, appears,
by the analysis of Bischof and Kjerulf, to be a doleritic rock. (H. von Dechen,
S. 383, 386, and 393.) The rock of the little Rosenau, which has sometimes
been called Sanidophyr, belongs, according to G. Rose, to the first division of his
trachytes, and is very nearly related to trachytes of the Ponza Islands. The
trachyte of the Drachenfels, with large crystals of glassy felspar, appears to
resemble most nearly (according to Abich's observations, which unfortunately
are not yet published) that of Dsyndserly dagh, 8500 feet high, which rises
to the north of the great Ararat, out of a nummulite formation, having under-
lying Devonian strata.

(W8) p. 431. — The near vicinity of Cape Perdica, on the island of JEgiua, to
the brownish-red Trcezene-trachytes of the peninsula of Methana (Kosmos, Bd.
iv. S. 273, Anm. 86 ; English edition, p. 227, and Note 310), and to the sulphur-
springs of Bromolimni, renders it probable that the trachytes of Methana, as
well as those of the island of Kalauria, near the little town of Poros, belong to
the same third division of G. Rose (oligoclase, with hornblende and mica).
Kurtius, Peloponnesos, Bd. ii. S. 439 and 446, Tab. xiv.

(599) p. 431. — See the excellent geological map of the country round Schem-
nitz, by Bergrath Johann von Peltko, 1852, and the Abhandl. der k. k. geolo-
gischen Reichsanstalt, Bd. ii. 1855, Abth. i. S. 3.

C"0) p. 431. — Kosmos, Bd. iv. S. 427, Anm. 7 ; English edition, p. 386,
and Note 531.

(M1) p. 431. — The basaltic columns of Pisoje, the felspathic part of which
has been analysed by Francis (Poggend. Annalen, Bd. lii. 1841, S. 471), and
which are situated near the bank of the Cauca River, in the plains of Amolanga
(not far from the Pueblos de Sta. Barbara and Marmato), consist of a somewhat
altered oligoclase in large fine crystals, and small crystals of hornblende. Nearly
allied to this mixture are : the quartz-containing diorite-porphyry of Marmato,
brought home by Degenhardt, and in which Abich called the felspathic constituent
Andesin ; the rock without quartz of Cucurusape, near Marmato, from Boussin-
gault's collection (Charles Sainte-Claire Deville, Etudes de Lithologie, p. 29) ;
the rock which I found in situ twelve geographical miles east of Chimborazo,
under the ruins of Old Riobamba (Humboldt, Kleinere Schriften, Bd. i. S. 161) ;

NOTES. civil

and lastly, the rock of the Esterel Mountains, in the Depart, du Var (Elie de
Beaumont, Explic. de la Carte ge'ol. de France, t. i. p. 473).

C602) p. 432. — The felspar in the trachytes of TenerifFe was first recognised
in 1842, by Charles Deville, who visited the Canary Islands in the autumn of
that year. See that distinguished geologist's Voyage geologique aux Antilles
et aux lies de Te'ne'riffe et de Fogo, 1848, p. 14, 74, and 169 ; and his Analyse
du Feldspath de Te'ne'riffe in the Comptes rendus de 1'Acad. des Sc. t. xix. 1844,
p. 46. He says, " Les travaux de Messrs. Gustav Eose et H. Abich n'ont pas
peu contribue sous le double point de vue crystallographique et chimique, a re'-
pandre du jour sur les nombreuses variete's de mine'raux qui etaient comprises
sous la vague denomination de feldspath. J'ai pu soumettre a 1'analyse des
cristaux isoles avec soin et dont la densite en divers echantillons etait tres-uni-
forme'ment, 2*593, 2'594, et 2'586. C'est la premiere fois que le feldspath
oligoclase a etc' indique dans les terrains volcaniques, a 1'exception peut-Stre de
quelques-unes des grandes masses de la Cordillere des Andes. II n'avait e'te
signale', au moins d'une maniere certaine, que dans les roches e'ruptives auciennes
(plutoniques, granites, sye'nites, porphyres sydnitiques ....); mais dans les
trachytes du Pic de TenerifFe il joue un role analogue a celui du labrador dans
les masses dole'ritiques de 1'Etna." Compare also Eammelsberg, in the Zeit-
schrift der deutschen geologischen Gesellschaft, Bd. v. 1853, S. 691, and the
4th Supplement to Handworterbuch der chemischen Mineralogie, S. 245.

C603) p. 432.— The first determination of the height of the great volcano of
Mexico, Popocatepetl, is, so far as I am aware, the trigonometric measurement
made by me on the 24th of January 1804, in the Llano de Tetimba. The
summit was found to be 1536 toises above the Llano, and as the latter is, ac-
cording to barometrical determination, 1234 toises above the coast of Vera Cruz,
we have the absolute height of the volcano 2770 toises, or 17,713 feet. The
barometric measurements which followed my trigonometrical determination, led
to the conjecture that the height of the volcano was still greater than that stated
by me in the Essai sur la Geographic des Plantes, 1807, p. 148, and in the
Essai politique sur la Nouv. Espagne, t. i. 1825, p. 185. William Glennie,
who, on the 20th of April 1827, was the first to reach the margin of the crater,
found the height, according to his own calculation (Gazeta del Sol, publ. en
Mexico, No. 1432), 17,884 English feet (2796 toises) ; or, according to a correc-
tion by Oberbergrath Burkart, who has done so much for American hypsometry,
and by comparison with an almost simultaneous barometric observation at Vera
Cruz, even 18,011 feet. A barometric measurement by Samuel Birkbeck (10th
Nov. 1827), calculated according to Oltmann's tables, again gave, however,
17,854 feet, and the measurement of Alexandre Doignon (Gumprecht, Zeitschrift

clviii NOTES.

fiir allg. Erdkunde, Bd. iv. 1855, S. 390), in almost too courteous agreement
with my Tetimba measurement, 5403 metres, =17,725 feet. The present ac-
complished Prussian Envoy at Washington, Herr von Gerolt, has also been on
the top of Popocatepetl, accompanied by Baron Gros (28th May 1833), and by
an exact barometric measurement found the height of the Roca del Fraile below
the crater 1 6,892 feet above the sea. A remarkable difference from the above-
named hypsometric results, which have been stated in chronological order, is pre-
sented by what appears to be a very carefully made barometric measurement by
M. Craveri, published in Petermann's valuable Mittheilungen iiber wichtige
neue Erforschungen der Geographic, 1856 (Heft x.), S. 358—361. This
traveller, in September 1855, found the elevation of the highest, i. e. of the
north-westernmost crater-margin, as compared with what he considered to be the
mean atmospheric pressure at Vera Cruz, only 5230 metres, =16,099 French
feet ; being 521 such feet (g'o of the whole height measured) less than I had
found by trigonometric measurement half a century before. Craveri also con-
siders the height above the sea of the city of Mexico to be 184 French feet less
than it was made by Burkart and myself at very different periods (instead of
2277 metres, he estimates it at only 2217 metres). I have treated more fully
of these variations, in plus and minus, from the result of my trigonometric mea-
surement (which unfortunately has not even yet been followed by a second one);
in the above-named Journal of Dr. Petermann, S. 479 — 481. The 453 deter-
minations of elevation which I made from September 1799 to February 1804,
in Venezuela, on the forest-covered banks of the Orinoco, Rio de la Magdalena,
and Amazons ; in the Cordilleras of New Granada, Quito, and Peru ; and in
the tropical parts of Mexico ; which have all been recalculated by Professor
Oltmanns, uniformly according to the formula of Laplace with the coefficients of
Ramond, and which were published, in 1810, in my " Nivellement barometrique
et ge'ologique" (Recueil d'Observ. astronomiques, vol. i. p. 295 — 334), were
made without exception with Ramsden's cistern-barometers, " a niveau constant,"
and not at all with apparatus in which one inserts successively several fresh-
filled Torricellian tubes, nor with the instrument described by me in Lame'therie's
Journal de Physique, t. iv. p. 468, and which was only occasionally employed in
Germany and France in the years 1796 and 1797. I used similarly con-
structed Ramsden's portable cistern -barometers in 1805, in a journey through
Italy and Switzerland with Gay-Lussac ; we were both satisfied with their per-
formance, and a comparison with results obtained in the excellent investigations
of Julius Schmidt (Beschr. der Eruption des Vesuvs im Mai 1855, S. 114 —
116) has recently shown that we had reason to be so. As I never ascended the
summit of Popocatepetl, but measured it trigonometrically, it is the more diffi-

NOTES. Clix

cult to understand the grounds of the extraordinary objection made by Craven,
in Petermann's Geogr. Mittheil. Heft x. S. 359 : " that the height ascribed by
me to the mountain is too small, because, as I have myself stated, I had em-
ployed the method of setting up fresh-filled Torricellian tubes." The apparatus
with several tubes is not applicable in the open air, and least of all on the sum-
mit of a high mountain. It is one of the means to which, with the conveniences
that cities afford, an observer may have recourse at long intervals of time, when
he is uneasy as to the condition of his barometer. In this way, as a means of
verification, I have had recourse to it on a few very rare occasions only, but I
would still recommend it to travellers as warmly as I did in my Observ. astr.
(vol. i. p. 363 — 373) : " Comme il vaut mieux ne pas observer du tout que de
faire de mauvaises observations, on doit moins craindre de briser le barometre
que de le voir de'range*. Comme nous avons, M. Bonpland et moi, traverse
quatre fois les Cordilleres des Andes, les mesures qui nous inte'ressaient le plus
ont etc* repetees a differentes reprises; on est retourne aux endroits qui parais-
stiient douteux. On s'est servi de temps en temps de Tappareil de Mutis, dans
lequel on fait 1'expe'rience primitive de Torricelii, en appliquant successivement
trois ou quatre tubes fortement chauffes, remplis de mercure recemment bouilli
dans un creuset de gres. Lorsqu'on est sur de ne pas pouvoir remplacer les
tubes, il est peut-etre prudent de ne pas faire bouillir le mercure dans ces tubes
memes. C'est ainsi que j'ai trouve', dans des expediences faites conjointement
avec M. Lindner, professeur de chimie a 1'e'cole des mines du Mexique, la hauteur
de la colonne de mercure a Mexico, dans six tubes de —

259-7 lignes (ancien pied de Paris).

259-5

259-9 „ „

259-9

260-0 „ „ •

259-9

Les deux derniers tubes seuls avaient dte* purge's d'air an feu par M. Bellardoni,
inge'nieur d'instruments a Mexico. Comme I'exactitude de I'expe'rience de'pend
en partie de la proprete inte'neure des tubes vides, si faciles a transporter, il est
utile de les fermer hermetiquement a la lampe." As in mountainous districts
the angles of altitude for* trigonometrical measurements cannot be taken from
the sea-shore, and the measurements themselves are mixed, being in part baro-
metric (often they are so to a third, and from that to a half of their whole
amount), the determination of the elevation of the high plain in which the base
line is measured becomes of great importance. As corresponding barometric
observations at sea are rare, and in most cases too distant, travellers are only

Clx NOTES.

too much disposed to assume, as the mean atmospheric pressure on the high
plain and at the sea-shore, results inferred by them from observations made
during a few days only, and perhaps also at other seasons. " Dans la question
de savoir, si une mesure faite au moyen du barometre peut atteindre 1'exactitude
des ope'rations trigonome'triques, il ne s'agit que d'examiner, si dans un cas donne
les deux genres de mesures ont etc faites dans des circonstances e'galement favo-
rables, c'est-k-dire, en remplissant les conditions que la thdorie et une longue
expe'rience ont prescrites. Le gdometre redoute le jeu de refractions terrestres ;
le physicien doit craindre la distribution si ine'gale et peu simultane'e de la
temperature dans la coloune d'air, aux extremites de laquelle se trouvent places
les deux barometres. II est assez probable que pres de la surface de la terre le
de'croissement du calorique est plus lent qu'k de plus grandes e'le'vations ; et pour
connaitre avec pre'cision la densite' moyenne de toute la colonne d'air, il faudrait,
en s'e'levant dans un ballon, pouvoir examiner la tempe'rature de chaque tranche
ou couche d'air superpose'e." (Humboldt, Kecueil d'Observ. astron. vol. i. p. 138
and 371, in the Memoir on Eefraction and barometric Measurements.) While,
on the one hand, the barometric measurement of Messrs. Truqui and Craveri
gives only 17,158 feet for the height of Popocatepetl, and, on the other, Mr.
Glennie's makes it 17,884 feet, a newly published result, by a traveller who has
examined the environs of Mexico and the districts of Yucatan and Chiapa
(Gymnasial-Professor Carl Heller of Olmiitz), agrees with mine to within thirty-
two feet. (Compare my memoir entitled " Ueber die Hb'he des mexicanischen
Vulkans Popocatepetl," in Dr. Petermann's Mittheilungen aus Justus Perthes
geographischer Anstalt, 1856, S. 479—481.)

(604) p. 432. — In the Chimborazo-rock it is not possible to separate mecha-
nically the crystals of felspar from the ground-mass in which they are embedded,
as can be done in the Etna-rock; but the proportionally large amount of silicic
acid contained, and the lower specific gravity of the rock connected therewith,
enable the felspathic ingredient to be recognised as oligoclase. The amount of
silicic acid contained, and the specific weight, are, for the most part, in inverse
proportion : in oligoclase and labradorite respectively the first is 64 and 53 per
cent, while the latter is 2*66 and 2'71. Anorthite has, with only 44 per cent
of silicic acid, the great specific gravity of 2'76. In felspathic minerals, which
are also isomorphous, this inverse ratio, between the amount of silicic acid and
the specific gravity, does not hold good with different forms of crystals, as
Gustav Eose has remarked. Thus, for example, felspar and leucite have the
same constituents — potash, alumina, and silicic acid; but felspar has 65, and
leucite only 56 per cent of^silicic acid ; and yet the former has a higher specific
gravity (2'56) than the latter (2'48).

As, in the spring of 1854, 1 wished for a new analysis of the Chimborazo

NOTES.

clxi

trachyte, Professor Rammelsberg was so kind as to undertake it, with his usual
precision. I subjoin the results as they were communicated to me by Gustav
Rose in a letter in the month of June 1854. " The piece of Chimborazo-rock,
which Professor Rammelsberg has submitted to a careful analysis, was broken
off from a specimen in your collection, which you had brought from the narrow
rocky crest at an elevation of 19,094 feet above the sea.

" ANALYSIS BY RAMMELSBERG.
(Elevation, 19,094 English feet; specific gravity, 2'806.)

Oxygen.

Silicic acid

59-12

3070

2-33

Alumina - - -

13-48

6-301

Protoxvde of iron

7-27

1-61 ^)

1

1 *ft

Lime ...
Talc

6-50
5-41

1-85
2-13 I

6-93J

Soda

3-46

0-89

Potash -

2-64

0-45 J

97-88

"ANALYSIS BY ABICH.
(Elevation, 16,178 English feet ; specific gravity, 2'685.)

Oxygen.

Silicic acid - - 65*09

33-81

2-68

Alumina - 15'58

7271

Oxyde of iron - - 3 -83
Protoxyde of iron - 173

1-16
0-39 1

Lime - 2'61

0-73 |>

1-0

Talc ... 4-10

1-58

Soda - - - 4-46

1-14 1

Potash ... 1-99

0-33 I

Loss ... 0-41

J

99-80

" (In explanation of the above numbers, it is to be remarked that the first co-
lumn gives the per-centage of the different constituents, and the second and third
the amount of oxygen in them. The second column indicates only the oxygen of
the stronger oxydes (which contain 1 atom of oxygen). In the third column
this is collected together, for the sake of comparison with the alumina (which is
a weak oxyde) and the silicic acid. The fourth column gives the proportion of
the oxygen of the silicic acid to the oxygen of all the bases taken together
as = 1. In the Chimborazo-trachyte this proportion is ~ 2'33 : 1-0.)
VOL. IV. I

clxii

NOTES.

" The differences between the analyses of Rammelsberg and Abich are, indeed,
rather considerable. Both the specimens of Chimborazo-rock analysed, one
from an elevation of 19,094 feet, and the other from 16,178 feet, were struck
off by yourself, and are from your geognostical collection in the Royal Cabinet of
Minerals in Berlin. The rock from the lesser elevation (scarcely 400 feet higher
than the summit of Mont Blanc), which was analysed by Abich, has a lower
specific gravity, and, in accordance therewith, a greater amount of silicic acid
than the rock from the higher elevation which was analysed by Rammelsberg.
Assuming the alumina to belong exclusively to the felspathic portion, we may
reckon in Rammelsberg's analysis : —

Oligoclase 58'66

Augite - - - 34-14

Silicic acid - 4'08

As with the assumption of oligoclase there thus still remains over some free
silicic acid, it is probable that the felspathic ingredient is oligoclase, and not
labradorite. This latter does not present itself with free silicic acid; and with
the assumption of labradorite in the rock, there would remain over still more
silicic acid."

A careful comparison of many analyses — for which I am indebted to the
friendship of M. Charles Sainte-Claire Deville, who had the free chemical use of
any part of the rich geological collections of our common friend Boussingault —
shows that the quantity of silicic acid, in the ground-mass of trachytic rocks, is
mostly greater thaji in the felspars which they contain. The subjoined table,
kindly communicated to me by the author, in the month of June 1857, includes
five of the great volcanoes of the Andes.

Names of the
Volcanoes.

Structure arid Colour of the Mass.

Silicic Acid in
the whole Mass.

Silicic Acid
in the Fel-
spar alone.

["half- vitrified, brownish grey

65-09 Abich "|

Chimborazo

< half-vitrified and black

63-19 Deville [•

58-26

L crystalline grey

62-66 Deville J

Antisana

("grey-black -

64-26 Abich "I
63-23 Abich j

58-26

Cotopaxi

f glassy and brownish -
j granular -

69-28 Abich 1
63-98 Abich J

Pichincha

black, glassy -

67-07 Abich

Purace'

almost bottle-green -

68-80 Deville

55-40

Guadeloupe

grey, granular, and cellular

57-95 Deville

54-25

Bourbon

crystalline, grey, porous

50-90 Deville

49-06

NOTES. clxiii

"Ces differences, quant a la richesse en silice, entrela pate et le feldspath,"
adds Charles Deville, " paraitront plus frappantes encore, si 1'on fart attention
qu'en analysant une roche en masse, on analyse, avec la pate -proprement dite,
non seulement des fragments de feldspath semblables a ceux que Ton en a extruits,
mais encore des mineraux qui, comme 1'amphibole, la pyroxene, et surtout le
peridot, sont moins riches en silice que le felspath. Get execs de silice se mani-
feste quelquefois par des grains isoles de quartz, comme M. Abich les a signale's
dans les trachytes du Drachenfels, et comme moi-meme j'ai eu 1'occasion de les
observer avec quelque e'tonnement dans le dolerite trachytique de la Guade-
loupe."

" If," says Gustav Rose, " we add to the remarkable table of the quantity of
silicic acid in the Chimborazo-rock the result of Rammelsberg's latest analysis
(May 1854), we find that Deville's result falls just intermediate between those
of Abich and Kammelsberg. We obtain : —

Chimborazo-rock.

Silicic acid 65'09 Abich (specific gravity, 2'685).
63-19 Deville.
62-66 Deville.
59-12 Rammelsberg (specific gravity, 2'806) "

In a newspaper appearing at San Francisco, in California, " 1'Echo d u Pa-
cifique," there is an account, on the 5th of January 1857, from a French tra-
veller, Monsieur Jules Remy, of his having succeeded, on the 3rd of November
1856, in company with an Englishman, Mr. Brenclday, m reaching the summit
of Chimborazo. He adds, " so wrapped in mist that as we mounted we were
not aware of it {sans nous en douter)." He observed the boiling point of
water at 77°"5 cent., the temperature of the air being 1°'7 cent., and on calcu-
lating the elevation from hence, " according to a hypsometric rule tested during
repeated journeys in the Sandwich Islands," he was surprised at the result he ob-
tained. He found that he had been at a height of 6543 metres (21,470 feet),
differing only forty French feet from that given for the top of Chimborazo by
my trigonometric measurement, taken in June 1803, from near Riobamba
Nuevo, in the high plain of Tapia. This agreement between a trigonometric
measurement, and one founded on the boiling point of water, is the more sur-
prising, because my result involved, as must always be the case in the Cordilleras,
a barometric portion (for the height of the plain of Tapia, 9484 feet), which for
want of corresponding observations on the sea-shore, could not have all the ac-
curacy that might be desired. (For details, see my Recueil d'Observ. astron.
vol. i. p. Ixxii. and Ixxiv.) Professor Poggendorff kindly undertook the trouble

12

clxiv NOTES.

of examining what would be the result given by a more satisfactory mode of
calculation under the most probable suppositions. Calculating under the two
hypotheses of the temperature of the air at the sea-shore having been either
27°'5 cent, or 26°'5 cent., and the height of the barometer reduced to the
freezing point 760mm 0, he found, according to Regnault's Tables, that 77°'5
cent, for the boiling point of water on the summit corresponds to a barometer
reading of 320mm 20 at 0° cent.;, the temperature of the air was + 1°7 cent.,
which we may take for convenience as 1°'5. According to these data, Oltmann's
tables give for the height attained, in the first hypothesis (27°'5 cent.), 7328ra
2, and in the second hypothesis (26°'5 cent.), 7314™ 5, or, on the mean, 777
metres more than my trigonometric measurement. To have agreed with this
last, the experiment, supposing the summit of Chimborazo to have been really
attained, should have given the boiling point 2°-25 cent, higher than it did.
(PoggendorfTs Annalen, Bd. 100, 1857, S. 479.)

(605) p. 433.— As early as 1833, when placing the rich Sicilian collections of
Friedrich Hoffmann in the Berlin Mineralogical Cabinet, Gustav Kose satisfied
himself and his friends that the trachytes of Etna contain labradorite. In the
Memoir on the Eocks designated as greenstone and greenstone-porphyry (Pog-
gend. Ann. Bd. 34, 1835, S. 29), Gustav Rose mentions Etna lavas which
contain augite and labradorite* (Compare also Abich, in the fine Memoir on
the whole felspar family, 1840, in Poggend. Ann. Bd. 50, S. 347.) Leopold
von Buch calls the Etna-rock analogous to the dolerite of the Basalt formation.
(Poggend. Bd. 37, 1836, S. 188.)

(606) p. 433. — Sartorius von Waltershausen, who has been for many years a
diligent explorer of the trachytes of Etna, makes the important remark: " That
hornblende there belongs rather to the older masses, the greenstone dikes in the
Val del Bove, and the white and reddish trachytes which form the base of Etna
in the Serra Giannicola. Black hornblendes and bright leek-green augites are
there found near together. The more recent lava-streams from 1 669 (particu-
larly those of 1787, 1809, 1811, 1819, 1832, 1838, and 1842) show augite,
but not hornblende, which appears to be produced under a slower rate of cool-
ing." (Waltershausen, Ueber die vulkani^chen Gesteine von Sicilien und Island,
1853, S. 11 1 —114.) In the augite-containing trachytes of the Fourth Division,
in the chain of the Andes, I have found augite abundant, but in some cases
absolutely no crystals of hornblende, and in some others (as on Cotopaxi at an
elevation of 14,068 feet, and on Rucu Pichincha at an elevation of 15,304 feet)
a few interspersed rare but distinct black crystals of hornblende.

(8cr) p. 433. — Compare Pilla, in the Comptes rendus de 1'Acad. des Sc.
t. xx. 1845, p. 324. In the leucite-crystals of the Eocca Monfina, Pilla found

NOTES. Clxv

the surface covered with worm-tubes (serpulese), which indicates submarine
volcanic origin. On the leucite-rock of the Eifel in the trachyte of the Burg-
berg near Rieden, and that of Albano, Lago Bracciano and Borghetto, north of
Rome, see Kosmos, Bd. iv. S. 32, Anm. 93 (English edition, Note 317). In
the centre of a large crystal of leucite, Leopold von Buch has often found the
fragment of a crystal of augite, around which the leucitic crystallisation has
taken place, "which, as has been already remarked, seems somewhat strange,
viewing the easy fusibility of augite and the non-fusibility of leucite. It is still
more frequent to find pieces of the paste of the leucite-porphyry itself enclosed
as a nucleus." Olivine is also found in lavas: as in the cavities of the obsidian
which I brought from Mexico from the Cerro del Jacal (Kosmos, Bd. i. S. 464,
Anm. 60); and yet it is also found in the hypersthene-rock of Elfdal, which
was long taken for syenite. (Berzelius, Sechster Jahresbericht, 1827, S. 302.)
Oligoclase presents a similar contrast in respect to the places in which it is
found: it occurs in the trachytes of still burning volcanoes (Peak of Teneriffe
and Cotopaxi) ; and yet it presents itself also in the granite and granitite of
Schreibersau and Warmbrunn in the Silesian Riesengebirge. (Gustav Rose, On
the Rocks belonging to the Granite Group, in the Zeitschrift der deutschen geol.
Gesellschaft zu Berlin, Bd. i. S. 364.) It is not so with leucite in Plutonic
rocks ; for the statement that leucites are found interspersed in the mica-slate
and gneiss of the Pyrenees at Gavarnie (which has even been repeated by Hauy),
has been found by Dufre'noy, by several years of local examination, to be erro-
neous. (Traite' de Mine'ralogie, t. iii. p. 399.)

(608) p. 435. — In a geological journey, which I made in 1795, through the
south of France, the west of Switzerland, and the north of Italy, I had been
convinced that the Jurassic limestone which Werner reckoned as belonging to
his Muschelkalk, constituted a distinct formation. In my memoir on subterra-
nean gases, which my brother, Wilhelm von Humboldt, published in 1799, while
I was in South America, this formation, which I had designated provisionally as
Jura limestone, was first mentioned (S. 39). It was thence immediately adopted
into Oberbergrath Karsten's " Mineralogische Tabellen," which were then much
read. (1800, S. 64, and Vorrede, S. vii.) I had not named any of the fossils
which characterise the formation, and respecting which Leopold von Buch did
such valuable service in 1 839 ; I was also mistaken in the age which I assigned
to it, as from the neighbourhood of the Alps, which were then believed to be
older than Zechstein, I supposed it to be older than the Muschelkalk. In the
earliest Tables of Buckland, " On the Superposition of Strata in the British
Islands," the " Jura-limestone " of Humboldt is reckoned as belonging to the

NOTES.

upper oolite. Compare my Essai ge'ogn. sur le Gisement des Roches, 1823,
p. 281.

(G09) p. 436. — The name of Andesite was first printed in a memoir of
Leopold von Buch's, read at the Berlin Academy on the 26th of March 1835.
As that great geologist restricted the name of trachyte to rocks in which glassy
felspar is contained, he said in his memoir read to the Academy in. March 1 835,
but not published until 1836 (Poggend. Ann. Bd. xxxvii. S. 188—190): —
" The discoveries of Gustav Eose on felspar have thrown a new light on volca-
noes and the whole of geognosy, and the kinds of rock in volcanoes have appeared
thereby in a new and entirely unexpected aspect. After many careful examina-
tions in the district about Catania, and on Etna, we, «'. e. Elie de Beaumont and
I, became convinced that there is no felspar in Etna, and therefore no trachyte.
All the streams of lava, as well as the beds in the interior of the mountain,
consist of a mixture of augite and labradorite. Another important difference in
the rocks of volcanoes appears when albite takes the place of felspar; a new
rock is thus constituted which can no longer be termed trachyte. According to
G. Rose's investigations (at that time), it may be affirmed almost decidedly that
not one of the numerous volcanoes of the Andes consists of trachyte; but that
they all contain, in the mass of which they are formed, albite. Such an asser-
tion seems a very bold one; but it will appear less so if we consider that almost
the half of these volcanoes, and their products in either hemisphere, are known
to us by Humboldt's travels alone. Further, this kind of rock, rich in albite, is
known through Meyen in Bolivia and the north of Chili, through Pb'ppig to the
extreme south of Chili5 as well as through Erman in the volcanoes of Kam-
tschatka. Such a marked and widely extended prevalence seems sufficient to
justify the name of andesite, by which this kind of rock, consisting of predomi-
nating albite mixed with a little hornblende, has been already more than once
presented to notice." Almost at the same time, in the additions with which, in
1836, he so materially enriched the French edition of his work on the Canaries,
Leopold von Buch entered into still further details; considering the volcanoes
Pichincha, Cotopaxi, Tungurahua, and Chimborazo all to consist of andesite;
and, on the other hand, calling the Mexican volcanoes true (sanidine-containing)
trachytes ! (Description physique des lies Canaries, 1836, p. 486, 487, 490.
and 515.) The lithological classification of the Mexican and Andes volcanoes
which has been given in this work, shows that there is no such uniformity of
inirieralogical constitution, and that no general designation, taken from an ex-
tensive region of the earth, can be scientifically applicable. Not long afterwards,
I also twice committed the fault of employing the name of andesite, the use of
which tends to create confusion: once, in 1836, in the description of my attempt

NOTES. clxvii

to reach the summit of Chimborazo, printed in Schumacher's Jahrbuch for 1837,
S. 204 and 205 (and reprinted in my Kleinere Schriften, Bd. i. S. 160 and 161);
and the second time, in 1837, in a memoir on the Highland of Quito, in Poggend.
Ann. Bd. xl. S. 165. I there said, having previously strongly opposed my
friend's statement that all the volcanoes of the Andes are similarly constituted : —
" Very recent times have shown that the different zones do not always present
the same (mineralogical) composition, — the same ingredients. Sometimes we have
trachytes proper characterised by glassy felspar, as at the Peak of Teneriffe and
the Siebengebirge near Bonn, where some albite is associated with the felspar:
these are felspar- trachytes, which, as active volcanoes, frequently produce
obsidian and pumice ; — sometimes melaphyres and doleritic compositions of
labradorite and augite, more nearly approaching basalt : as at Etna, Stromboli,
and Chimborazo; — sometimes albite with hornblende prevails: as in the newly
so-called Andesites of Chili, and the magnificent • columns of Pisoje, described
as diorite-porphyry, near Popayan, at the foot of the Volcano of Purace', or in
the Mexican Volcano of Jorullo; — and lastly, sometimes leucite-ophyrs, mix-
tures of leucite and augite: as in the Somma, the old wall of the crater of
elevation of Vesuvius." By an accidental misunderstanding of this passage,
which bears many traces of the imperfect state of knowledge at that time (still
assigning felspar instead of oligoclase to the Peak of Teneriffe, labradorite to
Chimborazo, and albite to the Volcano of Toluca), the ingenious investigator
Abich, himself both a chemist and a geologist, erroneously attributed to me (Pog-
gend. Ann. Bd. li. 1840, S. 523) the invention of the name andesite as that of a
trachytic, widely diffused, albite-containing rock; and gave to a new species of
felspar, first analysed by him, and respecting which there is still some obscurity,
the name of andesine, " in consideration of the rock (from Marmato near
Popayan) in which it was first recognised." Andesine (pseudo-albite in ande-
site) would be placed intermediately between labradorite and oligoclase : at the
temperature of 15° Reaumur its specific gravity is 2'733; that of the andesite,
in which the andesine was found, is 3' 593. Gustav Rose, and subsequently
Charles Deville (Etudes de Lithologie, p. 30), have doubted the independent
existence of andesine as a species : as it rests only on a single analysis of Abich's,
that made by Francis (Poggend. Ann. Bd. lii. 1841, S. 472), in the laboratory
of Heinrich Rose, of the felspathic portion of a fine piece of diorite-porphyry
which I brought from Pisoje near Popayan, indicating, indeed, great resemblance
to the andesine from Marmato analysed by Abich, but yet the composition being
different. There is still much greater uncertainty, as to the so-called andesine,
in the syenite of the Vosges (from the Ballon de Servance and from Coravillers,
analysed by Delesse). Compare G. Rose, in the already often-cited Zcitsclmft

clxviii NOTES.

der deutschen geologischen Gesellschaft, Bd. i. for 1849, S. 369. It is not un-
important to recall here that the name of andesine is first adduced by Abich, as
that of a simple mineral, in his comprehensive memoir entitled, " Beitrag zur
Kenntniss des Feldspaths" (Poggend. Ann. Bd. 1. S. 125 and 341; Bd. li. S.
519), in 1840, at least five years after the introduction of the word andesite, not
antecedently to it, as has been sometimes erroneously stated. The formations in
Chili, which Darwin so often calls " andesitic granite " and " andesitic porphyry,"
rich in albite (Geological Observations on South America, 1846, p. 174), may
very well also contain oligoclase. Gustav Rose, whose memoir " Ueber die Nomen-
clatur der mit dem Griinsteine und Grunsteinporphyr verwandten Gebirgsarten "
(in Poggend. Ann. Bd. xxxiv. S. 1 — 30) appeared in 1835, neither then nor
subsequently used the word " andesite," the definition of which, according to our
present knowledge of the nature of the ingredients, would be not " albite with
hornblende," but, in the Cordilleras of South America, " oligoclase with augite,"
The already antiquated story of this tern), which I have, I fear, treated at too
much length, concurs with many other examples in showing that, in the course
of the development of our physical knowledge, the descriptive sciences may gain
by th-3 stimulus to observation sometimes afforded by erroneous or inadequately
grounded distinctions (as in the tendency to reckon varieties as species).

(61°) p. 436. — As early as 1840, Abich described oligoclase-trachytes from
the summit-rock of Kasbegk, and from a part of Ararat (Abich, " Ueber die
Natur und die Zusammensetzung der Vulkan-Bildungen, S. 46); also, as early
as 1835, Gustav Rose (Poggend. Ann. Bd. xxxiv. S. 30) expressed himself
cautiously to the effect: " that hitherto in his determinations he had not taken
account of oligoclase and pericline, which yet are probably also components."
The belief which was formerly much entertained, that a decided predominance
of augite or of hornblende would also allow us to conclude as to a determinate
species in the felspathic series, — glassy orthoclase (sanidine), labradorite, or
oligoclase, — seems to be much shaken by a comparison of the Chimborazo- and
Toluca-rocks, of the trachytes of the fourth and third divisions. In basaltic
formations, hornblende and augite are often both equally abundant; that is by
no means the case in trachytes; but, in a very few instances, I have found
augite-crystals in Toluca-rock, and some crystals of hornblende in portions of
the Chimborazo-, Pichincha-, Purace'-, and Teneriffe-rock. Olivines, which are
so exceedingly rarely wanting in basalts, are as rarely to be found in trachytes
as they are in phonolites; and yet we sometimes see, in particular lava-currents,
olivines form in quantity by the side of augites. Mica is generally very unusual
in basalt; and yet some basaltic summits of the Bohemian Mittelgebirge, first
described by Reuss, Freiesieben, and myself, contain it in quantities. It is pro-

NOTES. Clxix

bable that the unusual sporadic occurrence of certain mineral bodies, and the
reasons of their regular or normal specific association, are dependent on the con-
currence of many not yet ascertained causes in respect of pressure, temperature,
character and degree of fluidity, and rapidity of cooling. Specific differences of
association are, however, of great importance in mixed rocks, as well as in dike-
or vein-masses ; and in geological descriptions drawn up in the open field, where
it has been possible to do so with the described object actually present, care
should be taken to distinguish whether any particular associated member is of
general prevalence, or at least is rarely absent, or whether its occurrence is only
occasional and apparently accidental. The diversity which is found in respect
to the elements of a composite rock (ex. gr. trachyte) is also to be observed,
as I have already remarked, in the occurrence of the rocks themselves. In
both continents there are large territories in which trachytes and basalts, as it
were, repel each other, as do basalts and phonolites; and other territories in
which trachytes and basalts alternate in considerable proximity to each other.
(Compare Jenzsch, Monographic der bb'hmischen Phonolithe, 1856, S. 1 — 7.)

(6:1) p. 437. — Compare in Bischof, Chemische und physikalische Geologic,
Bd. ii. 1851, S. 2288 and S. 2297; and Roth, Monographic des Vesuvs, 1857,
S. 305.

(612) p. 438. — Kosmos, Bd. iv. S. 365 (English edition, p. 320).

(613) p. 438. — It is scarcely necessary to remind the reader that the ex-
pression " is wanting " only implies that a mineral species has hitherto been
looked for in vain throughout a not inconsiderable portion of the volcanoes of an
extensive territory. I distinguish between : wanting (i. e. not found) ; rarely
found; and frequently present, but yel; not normally and as a characteristic
component.

(6M) p. 438. — Carl von Oeynhausen, Erkl. der geogn. Karte des Lacher
Sees, 1847, S. 38.

(615) p. 438. — See the Bergmannisches Journal von Kohler und Hofmann,
5th year, Bd. i. (1792) S. 244, 251, and 265. Basalt, rich in mica, as at the
Gamayer Kuppe in the Bohemian Mittelgebirge, is a rarity. I visited this part
of the Bohemian mountains in 1792, in company with Carl Freiesleben, who
afterwards travelled with me in Switzerland, and who very materially influenced
my geognostic and mining education. Bischof doubts mica being ever produced
in a pyrogenous manner, and regards it as a product of metamorphic action
"par la voie humide;" see his Lehrbuch der chem. und physikal. Geologie,
Bd. ii. S. 1426 and 1439.

(816) p. 438. — Jenzsch, Beitrage zur Kenntniss der Phonolithe, in the Zeit-
echrift der deutschen geologischen Gesellschaft, Bd. vni. 1856, S. 36.

Clxx NOTES.

(617) p. 439. — Gustav Rose, Ueber die zur Granitgruppe gehorigen Gebirgs-
arten, in the above-mentioned Journal, Bd. i. 1849, S. 359.

(61S) p. 439. — The porphyries of Moran, Real del Monte, and Regla (the
latter celebrated on account of the great silver riches of the Veta Biscayna, and
the vicinity of the obsidians and pearl-stones of the Cerro del Jacal and Cerro de
las Navajas) are, like almost all the metalliferous porphyries of America, entirely
without quartz (on these phaenomena, and some completely analogous ones in
Hungary, see my Essai geognostique sur le Gisement des Roches, p. 179 — 188
and 190 — 193); but the porphyries of Acaguisotla, on the road from Acapulco
to Chilpanzingo, as well as those of Villalpando to the north of Guanaxuato,
which are traversed by auriferous veins, contain, together with sanidine, grains of
brownish quartz. As at the Cerro de las Navajas, and in the basalt and pearl-
stone-containing Valle de Santiago, along which one passes in going from Valla-
dolid to the Volcano of Jorullo, the enclosures of grains of obsidian and glassy
felspar in the volcanic rocks are, on the whole, rare, I was the more astonished
when, between Capula and Pazcuaro, and especially near Yurisapundaro, I saw-
all the ant-hills filled with shining grains of obsidian and sanidine. This was
in the month of September 1803. (Nivellement barometr. p. 327, No. 366;
and Essai ge'ognost. sur le Gisement des Roches, p. 356.) I wondered how it
could be possible for such small insects to carry these minerals from a great
distance. I have seen with lively pleasure that an ardent explorer, Jules Mar-
cou, has met elsewhere with something quite similar. He says : — II existe sur
les hauls plateaux des Montagues Rocheuses, surtout aux environs du Fort
Defiance (a 1'ouest du Mont Taylor), une espece de fourmis qui, au lieu de se
servir de fragmens de bois et de de'bris de vege'taux pour e'lever son edifice,
n'emploie que de petites pierres de la grosseur d'un grain de mai's. Son instinct
la porte a choisir les fragmens de pierres les plus brillants; aussi la fourmiliere
est-elle souvent remplis de grenats transparents magnifiques et de grains de
quarz tres-limpides." (Jules Marcou, Resume' explicatif d'une Carte geogn. des
Etats Unis, 1855, p. 3.)

In the present lavas of Vesuvius glassy felspar is very rare ; not so in the
older lavas (for example in those of the eruption of 1631), with leucite crystals.
Sanidine is also found abundantly, without any leucite, in the Arso-current of
Cremate, in Ischia, of the year 1301, which is not to be confounded with
the more ancient one described by Strabo, which is near Montagnone and Rotaro.
(Kosmos, Bd. iv. S. 340, Anm. 61, and S. 447 ; English edition, p. 260, Note
385, and p. 407.) Glassy felspar is not found in the trachytes of Cotopaxi, or
the other volcanoes of the Andes : nor does it appear in the subterranean pumice-

NOTES. clxxi

stone quarries at the foot of the Cotopaxi. What have been formerly described
in these as sanidine are crystals of oligoclase.

(M9) p. 440.— Roth, Monographic des Vesuvs, S. 267 and 382.

(C2°) p. 440. — See above, Anm. 82 ; English edition, Note 606 ; Rose, Eeise
nach dem Ural, Bd. ii. S. 369 ; Bischof, Chem. und physik. Geologie, Bd. ii. S.
528 — 571.

C521) p. 441. — Gilbert's Annalen der Physik. Bd. vi. 1800, S. 53 ; Bischof,
Geologie, Bd. ii. S. 2265—2303.

C532) p. 441. — The later lavas of Vesuvius do not contain any olivine any more
than any glassy felspar ; Roth, Mon. des Vesuvs, S. 139. The lava-stream of
the Peak of Teneriffe of 1804, which has been described by Viera and Glas, is,
according to Leopold von Buch, the only one which contains olivine. (Descr.
des lies Canaries, p. 207.) I have shown elsewhere (Examen critique de 1'His-
toire de la Ge'ographie, t. iii. p. 143 — 146) that the statement that the eruption
of 1704 was the first since the conquest of the islands at the end of the 15th
century is erroneous. Columbus, in his first voyage of discovery, in the nights
from the 21st to the 25th of August, when about paying a visit to Dona Beatrir.
de Bobadilla, in the Gran Canaria, saw the fiery eruption in Teneriffe. In his
Journal, under the heading " Jueves 9 de Agosto," which comprises intelligence
up to the 2nd of September, it is said, " Vieron salir gran fuego de la Sierra de
la Isla de Tenerife, que es muy alta en gran manera ;'• Navarrete, Col. de los
Viages de los Espanoles, t. i. p. 5. The above-named lady is not to be con-
founded with Dona Beatriz Henriquez of Cordova, the mother of his illegitimate
son (the learned Don Fernando Colon, his father's historian), whose pregnancy
in the year 1488 contributed so materially to retain Columbus in Spain, and
thus to occasion the New World to be discovered for Castillo and Leon, and not
for Portugal, France, or England. (Comp. my Exam. crit. t. iii. p. 350, 367.)

(623) p. 441.— Kosmos, Bd. iv. S. 276, English edition, p. 230.

(624) p. 441. — An important part of the rock-specimens collected by me
during my American expeditions has been sent to the Spanish Mineralogicai
Cabinet, to the King of Etruria, to England, and to France. I do not refer to
the geological and botanical collections possessed by my noble friend and fellow-
labourer Bonpland, by the doubly sacred right of personal collection and dis-
covery. Such a wide distribution of collected materials, when (by means of very
precise notice of the localities in which each was found) the preservation of
groups in geographical respects is not prevented, offers the advantage of facili-
tating the most varied and strict examination of the mineral species whose
habitual association characterises the rocks.

clxxii NOTES.

(62S; p. 442.— Humboldt, Kleinere Schriften, Bd. i. S. 139.

O526) p. 442. — The same, S. 202, and Kosmos, Bd. iv. S. 357, English
edition, p. 312.

(627) p. 442. — Humboldt, Kl. Schr. Bd. i. S. 344. I have also found much
olivine in the " tezontle" of the Cerro de Axusco, in Mexico. Is this " tezontle"
cellular lava or basaltic amygdaloid ? The Mexican word tetzontli means stone-
hair, from tetl, stone, and tzontli, hair.

(62S) p. 442. — Sartorius von Waltershausen, Physisch-geographische Skizze
von Island, S. 64.

(689) p. 442.— Berzelius, Sechster Jahresbericht, 1827, S. 392; GustavRose,
in Poggend. Ann. Bd. xxxiv. 1835, S. 14 ; Kosmos, Bd. i. S. 464.

(Sao) 4 42— Jenzsch, Phonolithe, 1856, S. 37; and Senft,in his important
•work, Classification der Felsarten, 1857, S. 187. Olivine is also found, accord-
ing to Scacchi, in the calcareous blocks of the Somma, together with mica and
augite. I call these remarkable masses emitted blocks, not lavas ; the Somma
has never poured forth any of the latter.

(s31) p. 442.-Poggend. Ann. Bd. xlix. 1840, S. 591, and Bd. Ixxxiv. S. 302 ;
Daubre'e, in the Annales des Mines, 4eme Se'rie, t. xix, 1851, p. 669.

C632) p. 443.— Kosmos, Bd. i. S. 136, and Bd. iii. S. 615 ; English edition,
vol. i. p. 121, and vol. iii. p. 444.

(633) p. 443.— The same, Bd. i. S. 465. English edition, Note 293.

C531) p. 443. — Humboldt, Voyage aux Regions Equinoxiales, t. i. p. 156 —
165. (Ed. in4to.)

(ess) p> 444 — Compare Kosmos, Bd. iv. S. 365 ; English edition, p. 320.

(636) p_ 444 — Scacchi, Osservazioni critiche sulla maniera come fu sepellita
1'antica Pompei, 1843, p. 10 : against the view put forward by Carmine Lippi,
and subsequently defended by Tondi, Tenore, Pilla, and Dufre'noy, that Pompeii
and Herculaneum were overwhelmed and buried, not by rapilli and ashes ejected
directly from the Somma, but through the agency of currents of water. Roth,
Monogr. des Vesuvs, 1857, S. 458. (Kosmos, Bd. iv. S. 449 ; English edition,
p. 409.)

(637) p< 445 — Nivellement barometr. in Humboldt, Observ. Astron. vol. i.
p. 305, No. 149.

(•") p. 445. — Kosmos, Bd. iv. S. 367 ; English edition, p. 323.

(639) p 445< — On the pumice hill of Tollo, which is still two days' journey
from the active volcano of Maypu, which latter never sent forth a morsel of such
pumice, see Meyen, Reise urn die Erde, Th. i. S. 338 and 358.

(64°) p. 445. — Poppig, Reise in Chile und Peru, Bd. i. S. 426.

(6I1) p. 445.— Compare Kosmos, Bd. iv. S. 417; English edition, p. 375.

NOTES. clxxiii

(612) p. 445. — Franz Junglmhn, Java, Bd. ii. S. 388 and 592.

(643) p. 446. — Leopold von Buch in the Abhandl. der Akademie der Wiss.
zu Berlin aus den J. 1812 — 1813 (Berlin, 1816), S. 128.

(W4) p. 446. — Theophrastus de Lapidibus, § 14 and 15 (Opera ed. Schneider,
t. i. 1818, p. 689, t. ii. p. 426, and t. iv. p. 551), says this of the Lipari-rock

(645) p. 446. — Rammelsberg, in Poggend. Annalen, Bd. 80, 1850, S. 464,
and Fourth Supplement to his Chemische Handworterbuch, S. 169 ; compare
also Bischof, Geologie, Bd. ii. S. 2224, 2232, and 2280.

(616) p. 447. — Kosmos, Bd. iv. S. 323, 354, 357—360, 366—368, and 377 ;
English edition, p. 289, 310, 313—316, 322—324, and 332. Eespecting par-
ticular cases in the geographical distribution of the pumices and obsidians in the
tropical zone of the New Continent, compare Humboldt, Essai ge"ognostique sur
le Gisement des Eoches dans les deux he'mispheres " 823, p. 340 — 342, and
344—347.

INDEX

TO

VOLUME IV. PART I.

ABICH, connection between earthquakes and thermal and gaseous springs, p. 175 ;
mud-volcanoes in the Caucasus, p. 207; Caucasian mountain-system, p. 340 ;
chemical analyses, Notes 604 and 610.

Africa, volcanoes of, p. 332.

Airy, density of the earth, p. 449 ; figure of the earth, p. 483, and Note 18.

Aleutian Islands, volcanoes of, p. 352, 411.

America, volcanoes of Central, p. 262, 308 ; of North-west, p. 388—403 ; of South,
p. 2C9— 278.

Amsterdam Island, p. 370, Note 503.

Andesite, p. 428, and Note 609.

Ansango, Volcan de, p. 312—314, 316.

Antilles, volcanoes in the, Note 555.

Antisana, Volcano of, p. 310-314, 316, Note 450.

Ants, shining mineral fragments carried by them to form their hills, Note 618.

Arabia, volcanic activity in, p. 338.

Arago, rotation magnetism, p. 70; magnetic observations, p. 449; connection of
magnetic disturbances with auroras, Note 134.

Ararat, p. 239 ; ascents of, p. 340; eruptions of, p. 341.

Arcs (meridional and parallel), measurements of, p. 21—23, Notes 7—9 ; latest
conclusions from, p. 482.

Arequipa, volcano, height of, discussed, Note 369.

Argaeus, Mount, Note 356.

Aristotle, references to, p. 9, 14.

Artesian wells, p. 35 — 38.

Ascension, Island of, p. 330.

Asia, volcanoes of "Western and central, p. 334 — 342 ; of eastern, 342—348 ; of
eastern and southern Asiatic islands, p. 349 — 361 and 361 — 367 ; mountain-
systems, volcanic activity and area of depression in central, p. 414—417.

Atlantic Ocean, volcanoes in islands in the, p. 328.

Attraction, discussion of molecular and mass attraction, p. 10, 17, and Note 2.

Augite, p. 439.

Aurora, p. 153—159 and 495, Notes 205—211 ; connection of, with magnetic dis-
turbances, Note 134.

Auvergne, extinct volcanoes and volcanic phenomena in, p. 236, 327, 414, Note
566.

Azores, geological conformation of the, Note 308.

Azufral, de Quindiu, p. 218—219.

INDEX.

Bache, p. 81.

Baily, density of the earth, p. 32 ; calculation of Foster's pendulum experiments,

p. 474.

Barlow, p. 76.
Beaufoy, p. 64.
Belcher, p. 77.
Bessel, figure of the earth, p. 21 and 483, Notes 4 and 5 ; correction of pendulum

results, p. 459, and Note 20.
Biot, pendulum experiments, p. 27 and 453.
Bolivia, volcanoes of, p. 276.

Boracic acid vapours, eruptions of, in Tuscany, p. 209, Notes 288 and 289.
Borda, p. 63, 94, and 118.
Borneo, volcanoes in, p. 362—364, Note 491.
Bourbon, Island of, p. 368.
Boussingault, view of cause of earthquakes, p. 170, and Notes 238 and 253;

chemical analyses, p. 219, 421 ; tropical thunderstorms, Note 579.
Bravais, p. 157, Note 213.

Broun, terrestrial magnetism, Notes 154 and 155.
Bunsen, chemical investigations into volcanic products, p. 419—420.
Buch (Leopold von), elevation-craters, p. 225 ; general view of the volcanoes of

the earth, p. 324.

Carbonic-acid gas, springs of,~p. 202.

Cascade Mountains, p. 399 — 400.

Caucasus, volcanic activity in the chain of the, p. 208, Note 285.

Celebes, volcanoes in, p. 365.

Celsius, early suggestion of simultaneous magnetic observations, Note 72.

Central America, volcanoes in, p. 262.

Chemical analyses, by Rammelsberg, Abich, and Rose, Note 604.

Chili, volcanoes in, p. 277.

Chimaera, the, in Lycia, p. 252, Notes 289 and 375.

Chimborazo, p. 193, 315 ; recent ascent of, Note 604.

Chinese, knowledge of compass, p. 51 ; rope-boring, p. 216, and Note 298.

Christie, p. 76.

Clerk, p. 82.

Classification, hypsometric, of volcanoes, p. 245—248 ; of trachytes, p. 429.

Clouds, cirrus, supposed connection of, with aurora and magnetic direction, p.

156, Note 212.

Cofre de Perote, p. 306-307, and Note 443.
Colima, volcano, ascent and eruption of. Note 531.
Columbus, line of no magnetic declination, p. 55-58.
Comangillas, Aguas de (hot springs), p. 198.
Compass, early knowledge of the, p. 51—54, Note 54.
Conseguina (volcano), Note 390.
Cordilleras, their parallel branches, divisions, and nodes, Note, 402 ; snow-line

on the, Note 330.

Cotopaxi (volcano), p. 237, 318, 321, Notes 370, 454, and 458.
Craters, elevation-craters, description of, p. 225 and 228, Note 308 ; falling in of

walls of, p. 238 ; parasitic, near Vesuvius, p. 303.
Crystallised minerals in the " Maars" and in Vesuvius, p. 233.

Dana (James), his geological writings, p. 372 et seq., Notes 508 and 510—516.

Darwin (Charles), p. 278, 372 ; areas of subsidence, p. 371.

Davy (Sir Humphry), his theory of volcanic action, p. 169, Note 237.

INDEX. C

Dechen (Berghauptmann von), remarks on the maars, p. 231, and on the pumice,
p. 234, of the Eifel district, Note 314.

Decennial period of the solar spots and of the magnetic variations, pp. 81 and 89,
and Note 73.

Deception Island, p. 331.

Declination, magnetic, first marked on the chart of Andrea Bianco, 1436, p. 55 ;
line of no, determined by Columbus in 1492, p. 55—58 ; first map of, by Alonzo
de Santa Cruz in 1530, p. 56; existence of four lines of no, stated by Acosta
in 1589, p. 57; Halley's chart of (1702), p. 61; diurnal variation of, first
observed by Graham, in London, in 1722, p. 126; described, p. 127—135;
discussed, p. 496; present distribution of the, lines, p. 144.

Demavend, p. 335, Note 480.

Density of the earth, p. 31—33 and 449.

Deville (Charles St. Claire), p. 420, Notes 602, 604.

Diamaguetism, p. 50.

Diaphragm of Dicearchus, Note 567.

Disturbances, magnetic, p. 135, and 485—495.

Dome-shaped trachytic-mountains, p. 226.

Douglas (David), p. 73.

Due, p. 72.

Duperrey, p. 67 ; magnetic equator, p. 110.

Dupetit-Thouars, p. 67.

Earth, figure of, p. 18—31 and 453—484, Notes 7—10 ; density of, p. 31 and 449,
and Notes 25—28 ; temperature of, p. 34—48 and Notes 31—50; internal heat
of, Note 234 ; ideas of the ancients on the form of, Note 23 ; thickness of
crust of, p. 418, Notes 234 and 574.

Earthquakes, p. 166—184 and Notes 230—254 ; nature of the first impulse in
earthquake movements, p. 169; velocity of propagation of, p. 179; question
of whether they affect the condition of terrestrial magnetism, p. 123 — 125 ;
Riobamba, p. 172—173; Lisbon, p. 179, Note 251.

Ehrenberg, infusoria in the Eifel, p. 235—236.

Eifel, maars and ancient volcanoes in the, p. 229—236, 425, and 445, Notes
314—319, 511.

Elburuz, p. 341.

Elevation-craters — see Craters.

Elliot, magnetic determinations, p. Ill, 149.

Ellipticity of the earth, p. 18-31 and 453—484.

Encke, on the diurnal variation of the Declination, Note 154.

Erebus, Mount, p. 247 and Note 355.

Erman, magnetic observations, p. 72, 95, 100, 125, and 148—149 j volcanoes in
Kamtschatka, p. 343—348, and Notes 349 and 352.

Etna, p. 223, 243, 250, 259, and 425, Notes 605 and 606.

Eruption, cones of, p. 302.

Eruptions, remarkable submarine volcanic, p. 215 and 331.

Etymological discussions on the Pithecusae, Note 385.

Euboea, earthquake and emission of lava in the Island of, p. 182 and 224.

Faraday, on the magnetism of oxygen, p. 50 and 87 ; magnetic induction currents,
p. 75, 83, 91, 105, and Note 75.

Felspar, glassy, p. 439.

Fissures, pouring forth of lava and basalts from, p. 223, 259, 308, 384, and Note
306; volcanoes elevated over, p. 259; Italian, p. 260; American, p. 261
—264, 268-269, 276, 289, 299, 388 j Java and Sumatra, p. 279, 282 ; continent

VOL. IV. m

clxxviii INDEX.

of Asia, p. 340; Kamschatka and Aleutian Islands, p. 350, 353—354, 388;
Pacific Ocean, p. 373 and 382 ; New Zealand, p. 380.

Fitz Roy, p. 75.

Fogo : Volcano of the Ilha do Fogo, p. 257, 329 and Note 351.

Force, magnetic, points of greatest, p. 92, 95 ; line of least, p. 92, 101, 505 ; de-
crease in ascending above the earth's surface, p. 102 — 105 ; measures of its
variations first suggested by Borda, p. 63; first executed by La Perouse's
expedition, 1785 — 1787, p. 63; its variations first made known by Huinboldt,
p. 63; Sabine's map of, p. 77 ; diurnal variation of, p. 106 and 112—117.

Formosa, Island of, p. 350, 361, and Note 489.

Foster, magnetic observations, p. 105, 129 ; pendulum experiments, p. 472 — 479.

Franklin Bay, flames seen in, p. 482.

Fremont, the Rocky Mountains, p. 396.

Fumaroles, in Tuscany, the Caucasus, and Iceland, p. 209—211, and generally,
Note 555.

Fuss, p. 74.

Galapagos Islands, volcanic action in, p. 385 and Note 527.
Galera Zaml>a, submarine eruption at, p. 215 and Note 294.
Gaseous springs and exhalations, p. 206—221 ; Italian, p. 209 and 420, Note 289;

Icelandic, p. 210 and 420 ; American, p. 211 ; Chinese, p. 216 and 336; Javan,

p. 217; general remarks on volcanic gases, p. 420 — 421.
Gauss, terrestrial magnetism, p. 64, 75 and 86, 93, 485—487, 494.
Gay-Lussac, p. 104 and 118; earthquakes, Note 239.
Gay (Claudio), p. 81.
Geothermal lines, p. 189.
Gerolstein, p. 230.
Gey sir, p. 200, Note 274.
Gilliss, p. 82, Note 154 and 155.

Gilbert (William), on terrestrial magnetism, p. 58—59, and Note 59.
Gilpin, p. 64.

Graham, early observations on magnetic variations, p. 126, and Note 151.
Granite, fusion-temperature of, p. 244, and Note 464.
Greenwich magnetic observations, p. 83.
Grenelle, Artesian well of, p. 35.

Haidberg, magnetic rock of the, p. 160, and Note 224.

Halley, terrestrial magnetism, p. 60—61, Aurora, 135, and Note 167.

Hallmann, temperature of springs, p. 205, and Note 284.

Hansteen, terrestrial magnetism, p. 68, 69, 72, 119 ; movement of magnetic lines,

p. 152.

Hawaii, Island of, p. 374—377.
Heat, internal, of the earth, p 34—48, and Note 234.
Hecla, p. 241, 250, and 329.
Himalaya, p. 282.

Hopkins, on the depth of the solid crust of the globe, Note 234.
Hornblende, p. 439.

Hornitos, the, of Jorullo, p. 296-298, 302.

Ho-tsing, Chinese " fire-springs" of combustible gas, p. 216—217.
Humboldt, magnetic observations, p. 65-67, 71, 73, 77, 94-96, 103, 118—124,

and Notes 1 1 1 and 128.
Hypsometric tables, of volcanoes, p. 245—248 ; of Mexican volcanoes, p. 269 ; of

places in Mexico, p. 392 ; in the Rocky Mountains, p. 400—402.

INDEX.

Jan Mayen, Island of, p. 328.

Japan, volcanoes of, p. 355—361.

Java, springs of gas in, p. 217—218, volcanoes of, p. 279—289, 364 and 445, and

Notes 320 and 405 — 425; ribbed character, p. 284—285.
Jezo, Island of, p. 355.
Jorullo, description of, and circumstances attendant on its upheaval, p. 289—301,

and Notes 430, and 436.
Junghuhn (Franz), his writings on Java, p. 279—288 ; also, on volcanoes in other

islands in the Indian Ocean, p. 364—366.

Ice, stratum qf ground, p. 42—48, Notes 45—51 .

Iceland, its volcanoes and hot springs, &c., p. 210—211 and 329, 420, Note, 555.

Illimani, Note 372.

Inclination, magnetic, points of 90°, p. 92 and 108 ; line, of no, p. 92, 109 ; first
observed by Norman in 1576, p. 58; first map of, by Wilke, 1768, p. 63?
diurnal variation of, p. 106, 112-117; secular change of, p. 117—120; in-
fluence of height above, or depth beneath, the earth's surface on, p. 120-125.

Indian Ocean, volcanoes of, p. 367—371.

Intensity of magnetic force — see Force.

Ischia, p. 326, and Note 385.

Isodynamic, magnetic, lines, description, p. 95.

Isogonic, magnetic, lines, present distribution of, p. 144 — 153.

Izalco, Volcan de, p. 255—256.

Kamtschatka, volcanoes of, p. 342—348, Notes 349—352.

Karafuto, Island of, Note 485.

Keilhau, p. 71.

Kerguelen Island, p. 370.

Kilauea, lava-lake of, p. 375—376, and Notes 341, 512 and 513.

King, p. 71.

Kirauea— see Kilauea.

Kosima, volcano of, p. 244.

Kotzebue, p. 67.

Kreil, terrestrial magnetism, p. 81, 82, 84, and 451.

Krusenstern, p. 65 ; eruption of smoke in the Atlantic Ocean, p. 331 — 332.

Kupffer, ground-ice, p. 47 ; terrestrial magnetism, p. 73 and 74.

Kurile Islands, p. 354.

La Condamine, p. 275, 308, 309, and Note 446.

Lactagunga, pumice-stone quarries of, p. 320.

Lamanon, p. 63 and 94.

Lainont, terrestrial magnetism, p. 81 and 88, and Note 178.

Laplace, figure of the earth, p. 23—24 and 483.

Lava, emission of, from fissures, p. 182,223 — 224,308,348, Note 306; supposed
absence of, in Java, p. 288; and in the Cordilleras, p 309—316, and Notes 446
and 450; presence of, in some Central American volcanoes, p. 266; further
remarks, p. 288 ; ancient descriptions of, Note 306 ; angle of fall of streams
Of, p. 223.

Lava-fields, p. 304, 383, and Note 423.

Lava-lake — see Kilauea.

Let'roy, observations for the maximum of magnetic force, p. 96 and 97.

Lemnos, p. 326.

INDEX.

Leucites, p. 440 and Notes 317 and 607.
Lipari, p. 319.

Lloyd, terrestrial magnetism, p. 77 and 80.
Lutke, p. 71, 346—348.

Maars, in the Eifel, p. 229—236; in Auvergne, p. 236.

Macdonald, early observations on magnetic variations in Sumatra, p. 131 and
Note 162.

Madeira, p. 329.

Madre, Sierra, p. 394.

Magnetism, terrestrial, p. 49—161 ; early knowledge of, p. 51 ; ^chronological
enumeration of the advances in knowledge in the first half of the present
century, p. 65—85; insufficiency of the thermic hypothesis to explain the
observed phenomena, p. 87 ; connection with the solar spots, p. 81, 89, 492;
geographical distribution of the phenomena, p. 91 ; influence of height above
the earth's surface or of depth beneath it, p. 102 and 120, Notes 111 and 145 ;
magnetic variations according to hour and season, p. 106, 112, 126 and 496 —
516, and Notes 80 and 120; magnetic disturbances, p. 135 et seq.; and 485—
495 ; solar-diurnal variation, p. 496—516 ; lunar-diurnal variation, p. 451.

Makerstoun, magnetic observations at, p. 82.

Mallet on earthquakes, Notes 230, 234, and 239.

Malpais, of Jorullo, p. 293—295, 304-307, 383, 398 ; Note 423.

Masaya, Volcan de, p. 253, 255, Notes 376-378.

Mauna Loa or Mauna Roa, description of, p. 374—377, and Notes 359, and 511.

Methana or Methone, p. 227, 327, and Note 310.

Mexico, volcanoes of, p. 267—269, 304—307 and 387—388 ; broad elevated tract of,
p. 390.

Mica, p. 438.

Middendorff, ground-ice, p. 43—47, Notes 46, 47, and 51. '

Mitscherlich, chemical analyses, p. 214 ; fusion of granite, p. 244 ; on the Eifel
eruptions, Note 317.

Monk Wearmouth, pit waters of, p. 37.

Moon, magnetic influence of, p. 90 and 451 ; surface of, p. 164—165, and Note 589.

Moore (Commander), p. 82.

Monte Nuovo, p. 303.

Morichini, p. 66.

Mosenberg, the, p. 230, 232, 425.

Moya, p. 168, 171, and 174.

Mud-volcanoes, of the Taman peninsula, p. 207, 209, and 213; of Turbaco, p.
211—216, and Notes 294 and 296 ; in islands in the Indian Ocean, Note 299.

Murchison (Sir Roderick), eruption of vapours in Tuscany, p. 210, and Note 289 ;
effects of volcanic action in Llandilo flags, p. 328 and 448.

Naphtha springs, p. 206 and 207, 220, 227—228, Notes 304, 572.
Neu Salzwerk, Artesian well of, p. 36.
Nisyros, p. 327.
Nuovo, Monte, p. 303.

Obsidian, p. 443.

Oersted, p. 69.

Oligoclase, Notes 593 and 594.

Olivine, p. 441.

Orinoco, temperature of its waters, p. 186, and Note 256.

INDEX.

Orizaba, Volcano of, Note 366.
Owyhee, see Hawaii.

Oxygen, magnetic relations of, p. 50 and 91, and Note 75 ; quantity of, in rain
and spring water, Note 278.

Pacific Ocean, volcanoes in and around the, p. 372—386, 389, 411 ; arrangement of

islands in the, Note 508 ; nomenclature of parts of the, Notes 507 and 533.
Papagayos, storms so called, p. 264-265, and Note 389.
Paramagnetism, p. 50.
Patricius (Bishop), hot springs near Carthage and theory of internal terrestrial

heat, p. 197—198, and Note 269.

Pendulum experiments, p. 24—31, 453—484, Notes 14 and 19—22.
Perote — see Cofre de Perote.
Peru, volcanoes of, p. 276.

Phlegra or Phlegraean fields, p. 260 ; the Astroni in, Note 589.
Picard, early inquiries on the figure of the earth, p. 25.
Pichincha, p. 239—241.
Plana, on the density of the earth, p. 33.
Pompeii, causes and nature of the catastrophe, p. 409 and 444.
Popocatepetl, Note 603.
Pozzuoli, Note 555.
Pumice, in the Eifel, p. 234 ; Lipari, p. 319 ; Lactagunga, &c. p. 320—323, 408

—410, 443—447.
Pyriphlegethon, p. 34, 261.

Quartz its presence near South American volcanoes, p. 424.

Quetelet, p. 71.

Quito, volcanoes of, p. 274—276, 308.

Ram, influence of, on the temperature of variable springs, p. 186.

Reich, density of the earth, p. 31 and 32.

Ribbed character of some volcanoes, p. 284.

Richer, pendulum, p. 25.

Riobamba earthquake, p. 167, 172.

Rocky Mountains, p. 395 — 402, and Notes, 536—549 ; tables of heights and vol-
canoes in the, p. 400—402.

Rose (Gustav.) chemical analyses and classification of volcanic minerals, p. 426
et seq., and Note 610.

Ross (Sir James Clerk), magnetic observations, p. 77, 80, 108, 120, 148, and Notes,
88 and 97 ; Arctic voyages, p. 68, Antarctic voyage, p. 109, 370, and 411.

Ruidos, p. 177-178.

Sabine, magnetic observations, p. 68, 70, 71, 80, 109, 119, 129; magnetic discus-
sions, p. 72, 75, 76, 77, 78—80, 81, 82, 85, 87—90, 95—102, 106—107, 113—
117, 132—135, 139—143, 147—148, 451 and 485—516 ; Notes, 80, 81, 120, 154, 165,
166, 175 ; pendulum experiments, p. 27 and 453 — 484.

Sahama, p. 248.

St. Helena, Island of, p. 330.

St. Paul, Island of, p. 369—370.

St. Vincent, Volcano of, p. 177, and Notes 254 and 555.

Santa Lucia, Volcano of the Island of, Note 555.

Salses, p. 166, 206—221.

San Bruno, obelisks of, p. 171.

clxxxii INDEX.

Sang-ay, ascent of, by Wisse, p. 182 ; description, p. 257—259, 424, Notes 364, 383,
and ronquidos of, Note 383.

Schergin shaft near Jakutzk, depth of icy stratum in, p. 44, and Notes 45 and 46.

Schlagintweit (the Brothers), temperature of springs in the Alps, p. 192—193;
observations in the Kuenlun mountains, Note 572.

Schmidt (Julius), on Vesuvius, p. 347, and Notes 421 and 589.

Schomburgk, p. 82.

Schwabe, sun's spots, p. 89.

Senarmont, formation of minerals in veins, &c., p. 205, Note 283.

Siebold, description of Japanese volcanoes, p. 355 et seq.

Simonoff, p. 83.

Smell, sweet or peculiar, in some cases of volcanic action, p. 227—228.

Snow-line on the Cordilleras, Note 330.

Solfataras, p. 328, 381, 408, and Notes 519, 520 and 555.

Somerville (Mrs.), p. 67 and Note 88.

Soufriere of la Guadeloupe, Note 555.

Springs, thermal, 184—206, of vapour and gas, 206—221, Notes 289-300; tem-
perature of, p. 186, 271—276, Notes 256, 282, 284 ; mineral veins formed by,
p. 204, Note 283.

Strokr, p. 200.

Storms, maffnetic — see Disturbances.

Stioiiiboli, p. 251—252.

Styx, source of the, Note 282.

Sulphur-springs, p. 203.

Sun, its connection with terrestrial magnetism, p. 81, 88, and 492, Note 73.

Taal volcano in the Island of Luzon, p. 241, and Note 337.

Tahiti, p. 383—384.

Taman, mud-volcanoes of, p. 207, 209.

Temperature, distribution of, in the crust of the globe, p. 34—48, 188—197 ; in the

atmosphere, p. 188 ; in the ocean, p. 195; in Artesian wells, p. 35 ; stratum

of invariable, p. 38 and 189, and Note 41 ; of springs,p. 185—187, 189—190;

in the Alps, 192 ; in Germany and Italy, p. 205—206, and Note 284 ; of hot

springs, p. 198 ; of melting point of granite, Note 464.
Teneriffe, height of the Peak of, Note 354.
Thermal springs, p. 184—206.
Thian-schan, volcanic activity in the, p. 337 and 414; connection with the

Caucasus, p. 340, Note 567.
Tolima, Note 368.

Toluca, ascent of, by Humboldt, p. 387, and Note 531.
Trachytes, classification of, p. 429—436.

Trincheras, change of temperature in the hot springs of las, p. 199.
Turbaco, volcancitos of, p. 211—215, Notes 294, 295.

Vapour-springs, p. 206-221.

Vesuvius, p. 181, 346,407—410, 445, Notes 557, 558, 588.

Vidal (Captain), his descriptions and charts of the Azores, Note 308.

Vinagre, Rio, p. 203.

Vitruvius, quotations from, p. 408.

Volcancitos, p. 211.

Volcanic action, different theories of, p. 169, and Notes 235—237 ; ancient descrip-
tions of and ideas respecting, p. 224, 227, 260, and. Note 386 ; why more fre-
quent near seas, p. 411—417.

INDEX. clxxxiii

Volcanic formations, origin of some erupted rocks possibly anterior to the up-
heaval of volcanoes proper, p. 222—236 and 447—448.

Volcanism, or reaction of the interior of the earth on its exterior, p. 162 — 448.

Volcano, one of the Lipari Islands, p. 245, 326, Notes p. 342.

Volcanoes, general remarks, p. 221-228, 242, 248, 309, 323, 406, and 411 ; different
forms of the framework of, p. 236, elevation of, over fissures, p. 259, (and
see "Fissures;") comparative frequency on islands and coasts, p. 411; dis-
tances from the sea, p. 413—417; sounds emitted by, or "voices" of,
particular, p. 257, and Note 383 ; ribbed character in some, p. 284, and Note
421; five groups of, in America, p. 273 ; of Central America, p. 262 — 267,308;
and Notes 390, 391, and 396; of Mexico, p. 267—269, 289—307, 387, and 435,
and Notes 394, 395 ; of Quito and New Granada, p. 269, 274, and Note
397 ; of Peru and Bolivia, p. 276, and Note 398 ; of Chili, p. 277, and Note
399; of Java, p. 279, Notes 410—424; of Europe, p. 326 ; of islands in the
Atlantic, p. 328; of Africa, p. 332; of West and Central Asia, p. 334; of
North-east Asia, p. 342; of East Asiatic islands, p. 349; of Japan, 355; of
South Asiatic islands, p. 361 ; of the Indian Ocean, p. 367; of the Pacific, p.
372 ; of New Zealand, p. 380 ; of North America, p. 388 ; of the West Indies
or Antilles, Note 555 ; number of, on the globe, p. 406.

Whipple (Lt.), his expedition in the Rocky Mountains, p. 397.
Wrolf, on solar spots, p. 81, Note 73.

Younghusband, magnetic disturbances, p. 140—141.
Zealand, New, volcanoes and volcanic action in, p. 380—382.

END OP VOL. IV. PART I.

POINTED BT SPOTTI.SWOODS AKD CO.
NBW-STEEB'I 8QUAJ££.

•

THIS BOOK IS DUE ON THE LAST DATE
STAMPED BELOW

AN INITIAL FINE OF 25 CENTS

WILL BE ASSESSED FCfR-FAILURE TO RETURN
THIS BOOK ON THE DATE DUE. THE PENALTY
WILL INCREASE TO SO CENTS ON THE FOURTH
DAY AND TO $!.OO ON THE SEVENTH DAY
OVERDUE.

JULS 78

"** *3 J947

LIBRARY USE

LIBRARY USE

MAY 9 1955

— -

; . •

\

LD 21-95m-7,'37

YB 1 74 70

in
i f /

/ 6

.. ,

98833