ON

EE z
AO Sits

AE wens SEEN 3

Së

Xx PP

H
0
V

4
IK?

Wit

CS

A

0 - P - t -
۲ ^ , d E 4 e
E 7 D y Ki , 5
D . h
E
: 2
, 3
= ` 4 - D
e E >
7 d
٠ H
: ۰ ۰
" > R ٠
B s ` : ۱
D
E . D E A d e
* P D
` 22517 D a "
. . : لک‎ M -
3 4
7 4 ۰ E "
H B
` A
3 e
s
n °- ?
`
d
T 3
v ۹
«
Y Y za y å
¢ E
e š ; D
۰
; r d
E H d
e
- 4
;
e * g H
LI y E
1
'
» LI ,
۱ ۸ i
H
e E d
‘
=
` '
1
d ہک‎ an
e .
» ۹ Y
`
* g H
"
‘
M D
d à
٦ 1 2 - ,
J ' ۰ ور‎
>
' e y *
x AN
D A
, " D
è D
e e D »
. « " t Lë
+ d وہ‎ a * `
D A >. Ë " D "
` ب‎ f B
, i P. s '
‘ d ae A?
] "e وو‎
1 v
' A -
]
e ^
^ x

ai:
M.

Ert.

GEOLOGICAL SURVEY OF INDIA.

R_D.Oidham Memoirs, Vol XXTX Frontispiece.

Photo-etchmé§ - Survey of India Offices,Calcutta,July 1898.

MONUMENT TO GEORGE INGLIS, CHATAK.

(The part left standing is about 45 feet high.)

MEMOIRS

OF THE.

GEOLOGICAL SURVEY OF INDIA.

VoL. XXIX.

Published by order of His Excellency the Governor: General of India:
in Council.

81

CALCUTTA:
SOLD AT THE OFFICE OF THE GEOLOGICAL SURVEY.
LONDON, MESSRS. KEGAN PAUL, TRENCH, TRUBNER & Ce:

MDCCCXGCIX,

“HASTINGS s STREET. EN‏ و

s u TR : i be a | | کو‎ | e Ze "
os ^N 7 E
AN UA EN

CONTENTS

~ REPORT oz the GREAT EARTHQUAKE of 12TH JUNE 1897, dy
R. D. OLDHAM, A.B.S.M., F.G.S., Superintendent, Geological
Survey of India.

Preface. On some general principles of Seismology MET ۰ : v
Chapter I.—Introductory e > . E . ۰ . 1
Chapter 11.—Narrative accounts of the nature of the shock . . ۰ 4

Chapter III.—The isoseismic lines and area over which the shock was felt 42
Chapter IV.— The rate of propagation, and time of the commencement

of the shock e e S . . ° 53
Chapter V.—The rate and range of motion of the wave panicle . ex ETE
Chapter VI.— Earth fissures, sand vents and allied phenomena è e 85

Chapter VII.—The landslips . e ۰ ٠ . . . AE,
Chapter VIII.—The aftershocks ` : ` ° - Sek C a FIZA

Chapter 1X.—Results of an exploration of the epicentral tract . . . 09
Chapter X.—The position and extent of the seismic focus : ei “104
Chapter XI.—The records of the Bombay Magnetic Observatory . «| 06
Chapter XII.— Electric effects ` e . " e - . de
Chapter XIII.—The earthquake sounds, with some remarks on the Barisal

guns. : e . . . ; - e پر‎
Chapter XIV.—The rotation of pillars e monuments . . : 890ھ‎

Chapter XV.—The unfelt earthquake . s e . : ; 1 SP

Appendix A.—Reports by officers of the Geological Survey of India des-
patched to investigate the effects of the earthquake — .

(1) Report by Mr. T. D. LaTouche . . ° ا‎ 257

(2) Report by Mr. H. H. Hayden . . . 1 7ب‎

(3) Report by the late Mr. G. E. Grimes . : ee" 288

(4) Report by Mr. E. Vredenburg . : ; «15302

(5) Report by Mr. P. N. Bose . 4ء‎
(6) Further details from notes by Mr. R. D. Oldham 2> -7
Appendix B.—Details of fissures, sand vents and allied phenomena 0 319)

Appendix C.—Details of observations made for the purpose of de-
termining the velocity:of movement and acceleration of
the wave-particle, with a discussion of the formule em-

ployed and the results obtained . . : EC
Appendix D.—Details of the direction of overthrow and projection of free
; falling objects, with a consideration of the applicability
of this form of evidence to the determination of the

position of the epicentre . ; : : ‘ d eek

T CONTENTS.

Appendix E.— Details of the records of the seismometers at Shillong and
Silchar . ji : 3 38
Appendix F.—Details of the special HE made D the ××: 9
Department, referred, to in Chapter X e 3 «^ 359]
Appendix G.—Details and discussion of the results of the revisionary
triangulation of the Khasi Hills, carried out by the
Great Trigonometrical Survey in the season 1897-98 . 35E
Appendix H.—A review of the extent of the Lisbon earthquake of rst
November 1755, as compared with that of the great
earthquake of 12th June 1897 . " e . + 37%
Appendix 1.— Bibliography of the earthquake e S. AE, : e 371

LIST OF ILLUSTRATIONS.

Frontispiece. Monument to George Inglis at Chhatak..

Plate I. All Saints’ Church, Shillong.
Plate 11. Rookwood, Shillong.
Plate 111. Landslips at Beadon’s Falls, Shillong:
Plate IV. The Cemetery, Shillong.
Plate V. The Cemetery, Cherrapunji.
Piate VI. Bent rails at Rangapara, Tezpore—Balipara Tramway..
Plate VII. Manshai Bridge, Eastern Bengal State Railway. :
Plate VIII. Station yard, Nilphamari, Eastern Bengal State Railway.
Plate IX. Sand vents near Shaistaganj Station, Assam, Bengal Railway.
Plate X. Fissure at Rowmari.
Plate XI. Sand vent at Rowmari,
Plate XII. Khasia monuments at Lailang Kot.
Plate XIII. Fault crossing the road at Samin, Garo Hills.
Plate XIV. Waterfall formed by the Chedrang fault.
Plate XV. The Chedrang fault cutting the bed of the Chedrang River.
Plate XVI. Slope formed by the Chedrang fault near Dilma.
Plate XVII. Fig. 1. All Saints’ Church, Shillong, before the earthquake.
A 5 Fig. 2. Office of the Superintendent of Telegraphs, Shillong.,
Plate XVIII. Fig. 1. Government House, Shillong ; before the earthquake.
وو‎ » Fig. 2. Ruins of Government House, Shillong.
Plate XIX. Fig. 1. Quinton memorial, Shillong.
F 2 Fig.2. The same after the earthquake.
Plate XX. Fig. 1. Deputy Commissioner's cutcherry, Gauhati.
2 Le Fig. 2. Commissioner's cutcherry, Gauhati.
Plate XXI. Fig. 1. Tombs in old cemetery, Gauhati.
acea Fig. 2. Gate pillars of telegraph signallers’ quarters, Gauhati,
¿Plate XXII. Pet Goalpara Bazar from the east.
= "s Fig, 2. Fissure at east end of Goalpara Bazar.
Plate XXIII. Fig. 1. No. 9, Harrington Street, Calcutta.
Fig. 2. No. 6, Elysium Row, Calcutta.
Plate XXIV. Fig. 1, No. 114, Circular Road, Calcutta.
a S Fig. 2. House in Bentinck Street, Calcutta.

ete

CONTENTS, dp

Plate XXV. Fig. 1, Fissure in compound of Judge's house, Rangpur.
مار‎ Fig. 2. Collector’s house, Rangpur.
Plate XXVI. Fig. 1. Bridge over stream at Rangpur.
= E Fig. 2. Bridge over canal at Rangpur.
Plate XXVII. Fig. 1. Fissures in rock, at Balpakram, Garo Hills.
d 3s Fig. 2. Landslips in the Garo Hills, nerth of Balpakram,
Plate XXVIII. Fig, 1. Ruins at Muktagachha.
d J Fig. 2. House at Muktagachha.
Plate XXIX. Fig. 1. Godowns of jute mill, Sirajganj.
d Fig. 2. Manager's Bungalow, jute mill, Sirajganj.
Plate XXX. Fer Landslips near Cherrapunji.
Fig. 2. Landslips in Mahadeo Valley, Garo Hills.
Plate XXXI. Displaced boulders near Kanchi, Khasi Hills.
Piate XXXII. Fig. r. Tomb in Darjiling cemetery.
7 Fig. 2. Cracks in wall of jail, Maimansingh.
Plate XXXIII. William’s memorial, Shillong.
Plate XXXIV. Twisted gate pillars, Shillong.
Plate XXXV. Twisted pillars at Chhatak and Cherrapunji.
Plate XXXVI Tombs in old cemetery, Gauhati.
Plate XXXVII. Plans of twisted tombs and pillars at Gauhati.
Plate XX XVIII. Plans of disturbed tombs in Shillong cemetery.
Plate XXXIX. Hodograph, and Time-curve in East Indian Railway
between Howrah and Delhi.
Plate XL. Magnetograph records at Bombay,
Plate XLI. Records of the seismographs at Rocca de Papa and Catania,
Plate XLII. Rough plans of the Samin and Chedrang Faults.
Plate XLII. Triangulation chart of the Khasi Hitis.

MAPS,

Map No. 1.—The great Indian Earthquake of 12th June 1897 showing seismic
area, epicentre and isoseists.

Map No. 2.—Part of Assam to illustrate the examination of the epicentral tract
of the great earthquake of 12th June 1897.

Map No. 3.—Showing boundaries of the principal earthquakes of India during
the 19th century.

WOODCUTS IN TEXT.

Page,
Fig. r.-—lllustrates Capt. A. Howell's report . . . . ; d 10
Fig. 2.—Ditto Ditto ` s P . - 3 II
Fig. 3.—Typical section of a Calcutta house. . RT MIN : 44
Fig. 4.—Section of roof, cornice and parapet of a private house, Caleta 40
Fig. A ء5-‎ ء٥۱۱٥‎ of roof, cornice and parapet of Calcutta Town Hall : 46

Fig. 6.—Portion of trace of the barograph at Alipur on 12th June 1897,
showing the effect of the earthquake . . . : 5 58

105
111
131

EIS

166
173

214
216
218
221
222
260

261
262
264

272 -
274
275
282
303
304,
304
305
306
306

| 810

312
313
314
346
348
349

iv CONTENTS.
Fig. 7.—Section of tombs in the cemetery at Cherrapunji ۰ ‘ °
Fig. S.—Diagram to illustrate the formation of earth fissures . e e
Fig. 9.—Diagram to illustrate the formation of sand vents . . .
Fig. 10.— Diagram to illustrate the formation of fissures in alluvium . 5
Fig. 11.—Diagram to illustrate the displacement of alluvium along the foot
ofahil . : . e e E
Fig. 12.—Diagram to illustrate the i up of stream E and. exca-
vations . , . e » ۰ .
Fig. 13.— Diagram to illustrate the tir of landslips. . e e
Fig. ا‎ ates of Khasia monuments near Rambrai . . .
Fig. 15.— View of the Arbela range, from the road to Tura. e s
Fig. 16.—Sal tree in forest near Bordwar, broken by the earthquake . .
Fig. 17.—Section east from Quinaig, in the northern Highlands of
Scotland . : . - ۰ .
Fig. 18. —Schmidt's modification ER Se 5 facie . : . .
Fig. 19.—Diagram (after Mallet) to illustrate the effect of two shocks in
causing rotation . . e e e
Fig. 20.—Diagram to illustrate Gra y *s ec ‘i rotation + : : e
Fig. 21.—Diagram . ` : - - . E T e
Fig. 22.—Diagram of rotation by SE motion a e . e
Fig. 23.— Ditto Ditto . . ° . :
Fig. 24.— Plan of pillars of thermometer shed at Dhubri . . . .
Fig. 25.—O verthrow of plant-tubs at the Assistant Commissioner's bunga-
low, Goalpara . . e . . . e ٠
Fig. 26.—Tomb of Ensign Law, Gola کن‎ EE Ka . e
Fig. 27.—Plan of overthrown gate pillars at Gauhati e e
Fig. 28.—Plan of overturned gate pillars at entrance to its ids
Cherrapunji = . e s e " . . e
Fig. 29.—Overturned tomb in cemetery, Sylhet ۰ ٠ . . e
Fig. 30.—Hindu temple in cantonments, Sylhet e . e . :
Fig. 31.—Plan of twisting of obelisk in the cemetery, Darjiling . e e
Fig. 32.—Chimney in jail at Bhagalpur . . a a c e .
Fig. 33.— Watch tower on jail wall, Bhagalpur : e e een
Fig. 34.—Cross section of church at Bhagalpur . e . . .
Fig. 35.—Plan of roof trusses in church, Bhagalpur. ۰ . e e
Fig. 36.—House at jamalpur, facing S. 20° W. . e . .
Fig. 37.— House at Jamalpur, facing W. 20° N. : e : e °
Fig. 38.—Railway station at Tinpahar . : > E NL
Fig. 39.—Floring in Nawab's palace, Murdudatad" e e : -
Fig. 40.—Elevation of buildings at Berhampur : - 2 e .
Fig.*41.—Injury to roofs at Berhampur . : 5 e e : .
Fig. 42.— Diagram to illustrate overthrow S r e . = e
Fig. 43.— Ditto Ditto e ٠ . . e ٠
Fig. 44.— Ditto Diito e e . o . °

PREFACE.

ON SOME GENERAL PRINCIPLES OF SEISMOLOGY.

(The memoirs of the Geological Survey of India usually assume a certain
previous knowledge of the subject they treat of, and do not attempt to re-
capitulate the elementary principles which will be found in any text-book of geology.
In the case of the earthquake of 12th June 1897, which has attracted more general
attention than is usually given to the labours of the Geological Survey, more
elementary details than customary have been introduced, but even so there are
some points which it is necessary to understand, and of which a sufficiently exact
knowledge is by no means common. These could not have been treated in the
body of the report without great interruption to its continuity and the introduction
of a large amount of matter which would only embarrass those to whom it was
already familiar.

For this reason an introduction dealing with certain general principles has
been prepared and printed asa preface to the report. It is not intended as an
introduction to, Or as a primer of, seismology in general, but only of such parts of
the science as is necessary for the comprehension of the following report, or as have
not been treated with sufficient fullness in the body of the réport. Those who
desire to pursue the matter further should refer to the two volumes of the Inter-
national Science Series, * Earthquakes "and ‘ Seismology ? by Prof. J. Milne, F.R.S.,
which between them practically cover the whole range of the science.

A special index is appended to this preface to serve the purpose of a glossary
of the technical terms used in the report.)

J.—The nature of wave motion.

1. When a stone is dropped into a pool of water a series of concentric ripples
are formed, which travel outwards in ever widening circles till they reach the
edge of the pool, or, if it is large enough, gradually die out. But though we speak
of the ripples, or waves, as travelling, itis important to remember thatthe water
of which they are composed does not travel ; each separate particle of it moves
in a certain orbit, whose size and shape determines that of the wave, but ultimately '
returns toits original position, The travel of the wave is due to the fact that all
the particles of water do not move at the same time, each one starting and finishing
a little later than the one behind and a little earlier than the one in front of it, on the
course the wave is travelling.

2. The shape of the orbit in which the particles of water travel changes
gradually as the wave progresses, as may be seen in the gradually changing
form of the waves as they travel outwards, but the change is slow and, for any
given spot, the tracks in which all the particles of water move may be regarded
as similar, and the nature of the wave motion can be defined by the shape of
the track followed by each particle involved in the wave. Hence comes a term
which will be frequently employed, the wave particle: by this is meant no

H

E)

vi OLDHAM: GREAT EARTHQUAKE OF 1897.

specific individual particle of matter, but any ore taken at random at a chosen
place, by the extent, form, and rate of whose movement the nature of the wave
motion at that place is defined.

3. The particles referred to are molecules, the ultimate portions of matter
reached by continual subdivision till a limit is reached where no further division is
possible without a change of character and properties. Hence the movement in
a wave is called molecular, while a bodily displacement of the water or other
substance may be distinguished as molar. When a kingfisher dives into a pool it
makes a splash, and then from the splash a series of concentric ripples travel out-
wards in ever widening circles. In the splash the movement of the water is molar,
in the ripples it is molecular. So too the general movement of the water of a
trout stream may be regarded as molar, that in the ripples formed by the rising
trout as molecular. E

4. Gravitationa’ waves.— Waves are of several kinds; of these it will be well
t» take first that which is most commonly known, such as may be seen or started
on any sheet of water. Inthese the course of the wave-particle is elliptical, or
circular, as is diagrammatically represented in fig. i. The wave-particle is at
its highest point as the crest of the wave passes over it, and at its lowest when
the trough occupiesthe same position. When any one particle is at its highest
the ane in front has not yet risen to its highest point, while the one behind has
begun to sink again. By the time the one in front has reached its highest point
the one we were considering has begun to sink; and so it is that the wave travels.

Fig. i. Diagram of movement of wave particle in a gravitational wave.

5. Wave motion of this nature is quite superficial, and where waves'are set up
in deep water it is only that near the surface which is affected. As we descend
f rom the surface the extent of movement of the wave-particle rapidly diminishes,
the shape of its track becomes a more elongated ellipse, till, at a comparative short
distance from the surface, there is only a slight to and fro movement which
becomes less and less till it vanishes.

6. If in fig. i the firm and broken curved lines are taken torepresent the out-
lines of a wave when its crest and trough respectively are over the same place, and
the horizontal dotted line the normal level of the surface of the water ; thenthe
length ofthe wave isthe distance from crest to crest or trough totrough. The
Gistance a a' is the amplitude of the wave, or the distance the wave-particle moves
up or downwards from its normal level ; double this, or a' a' is the height of the
wave from trough to crest.

7. M the track of the wave-particle as drawn in fig. iis examined, it will be

seen that the movement can be divided into two phases : one of forward, and one of

PREFACE. vii

backward, movement, if the dividing line is horizontal, and one of upward, and one
of downward, movement if the dividing line is vertical. Moreover, by the inter-
section of these two dividing lines we get four semiphases in which the motion is
(1) upward and forward, (2) downward and forward, (3) downward and backward,
(4) upward and backwards, thus completing the cycle.

8. Waves of the kind now being dealt with are known as gravitational waves,
as in them the moving force is the attraction of the earth. The water which has
been raised above its normal level ina wave,is brought back by gravitation and
communicates the energy so developed to the water in front, by which it is
raised above its normal level, only to fall back and cause the rise of fresh
water m front. In other words, it is the weight of the water in a wave which
causes it to sink, and in doing so to make the water in front rise as the continua»
tion of the same wave.

9. Elastic waves.—Besides gravitational waves there are other molecular
movements which, by analogy, are called waves, and, being propagated in virtue of
the elasticity of the substance through which they travel, are called elastic waves.

The elasticity of a piece of indiarubber or cane is familiar enough; in the
sense used here it may be defined as the capacity, possessed by all substances to a
greater orless extent, of recovering their original bulk or shape, if these are
altered by the action of an external force. Itis of two kinds, one, common to both
solids and fluids, is the power of recovering the original volume, which has been
altered by change of pressure, when the original pressure is resumed. The measure
of this is the bulk modulus of the substance. The other kind of elasticity can only
be possessed by solids, and is the power of recovering the original form, if this is
changed by the action of an external force, when the force is removed, The
measure of this is known as the rigidity of the substance.

10. A perfectly elastic substance would completely recover its original bulk and
form, and in doing so give up exactly the same amount of energy as was absorbed
in producing the change. So far as bulk is concerned a gas may be regarded
as a perfectly elastic substance, but no solid can be regarded as such or even nearly
such, except within comparatively small limits. A bar of iron bent slightly will
spring back to its original form, but if bent further and strained beyond the limit of
elasticity will remain bent. So all solids, if strained within their limit of elasticity
will recover their original form, but if strained beyond that limit will either remain
permanently bent or break.

11. Within these limits elasticity is measured by the force required to produce
a given amount of change in bulk or shape. The greater the force that has to be
applied, the greater that which the substance will exert in regaining its original
bulk or shape, and the one is equal to the other. Measured in this way the elas-
ticity of steel or glass is very high, while that of the typical elastic band is very
small ; for, though the limit of elasticity is much wider, the force required to
produce a small change of shape is much smaller.

12. Waves of elastic compression.— Corresponding to the two kinds of elasticity
there are two kinds of elastic waves, one of which, depending on the resistance to
compression, can be propagated in solids and fluids alike. This form, which will
` be considered first, is that of those known as waves of elastic compression or more
briefly condensational waves. In these the movement of the wave-particle is

vii OLDHAM: GREAT EARTHQUAKE OF 1897.

directly forwards and backwards in the direction of travel of the waves, hence the y
are sometimes called normal waves.

13. To understand the nature of the movement in a wave of this kind let o, fig. نا‎

represent the normal position of

a molecule when undisturbed.

e o4 Suppose it to become involved in

a wave of elastic compression

travelling in a straight line from

Fig. Diagram to illustrate the successive posi- left to right ; then, starting from o

tions aries by the wave-particles at equal inter- „ : `
vals of time during one complete undulation of a it will move outwards passing

condensational wave, successively through the positions

1, 2, 3, tillat 4 it reaches the
extreme limit of its movement. From 4 it retürns with an ever-increasing
speed, till at 8 it passes on through its original position and, gradually slowing
down, comes to a stop at 12; from this it commences its return journey through
o to 4 and so on. The movement is in fact just like that of the bob of a pendulum;
assuming the direction of travel of the wave to be that of the swing of the
pendulum. Returning to the diagram fig ii, the distance from o to 4 is the
amplitude of the wave, or the extreme distance which the wave-particle reaches
from its normal position; 4 to 12 is the double amplitude or range of motion
of the particle, and the time taken by the wave-particle in travelling from o out to 4
and back through 12 to o is the period of the wave.

14. As in the water wave, the movement of the wave-particle may be divided into
two phases, one of movement in the direction the wave is travelling, the other in the
reverse direction. Each of these can again be divided into semiphases according
as the movement of the wave-particle is inwards, towards, or outwards from its
normal undisturbed position.

15. To understand what is meant by the wave length in a wave of this nature
we must consider how the arrangement of the molecules is affected by them. In
fig. iii, let o represent a molecule which at the movement is passing through its

e
6
7

DN
© O
KEE
e e tg

ôw
LJ
!
o
o
o
e
e
e
5
A
e
LJ
e
T
é
1
1
۱
©
۶
I
©
6

Fig. iii. Diagram to illustrate the distribution of matter in a condensational wave:

normal position ; the next molecule behind it, the wave being supposed as travelling
from left to right, is in a slightly more advanced stage of the wave motion, the next
still more so and so on till we come to one where the wave-particle has reached the
stage marked 2 in fig. ii, The small dot numbered 2 in fig. ii. may be taken to re-
present the normal position of the molecule and the heavy dot its position at the
instant under consideration. Passing still further backwards along the wave path
we come to molecules occupying respectively the positions 4,6,and 8, or normal
and behind that again 10, 12, 14, and so on to position 0. Now if the momentary
disposition of the dots is considered they will be seen to be crowded together at o
and widely separated at 8; that is to say there are definite zones in the substance
through which the wave is travelling where its material is alternately condensed
and rarified. We may regard the point of maximum condensation as analogous

PREFACE. ix

o the crest of a sea wave and the point of maximum rarefaction to the trough,
The wave length becomes then the distance o to o, or the length of the complete
undulation is analogous to the length of sea waves measured from crest to crest.

16. It isto be noticed that the points of maximum condensation or rarefaction
are those-of junction of the,semiphases of movement towards and away from the
normal position of the molecule; that of greatest condensation is where this point is
passed in the phase of motion in the same direction as the waveis travelling, that of
greatest rarefaction where the same point is being passed in the opposite phase of
motion. It may also be noticed that the tracts which are in this latter phase are
longer than those in the former, the difference being four times the amplitude.

17. As has already been remarked waves of this nature can be transmitted
through fluids as well as solids ; in the latter they travel faster than any other
form of elastic wave. It is waves of this nature which, if the period is not too small
or too great, become sensible as sound,

18. Waves of elastic distortion.—Another form of elastic waves is due to the
rigidity of the substance through which it is
transmitted, or to that form of elasticity which
causes a substance to resist a change of itsoriginal

e shape, by the action of an external force, and to
regain it when this is removed. These waves,
which are necessarily confined to solids, and
cannot be transmitted by fluids, are known as
ger waves of elastic distortion or distortional waves.
In them the motion of the wave-particle is still in
a straight line backwards and forwards, but this
line lies at right angles to, instead of along, the
direction in which the wave path is travelling, the
displacement being alternately to one side and the
other of the normal position of the wave-particle ;
74 ® 0 hence these waves are sometimes distinguished as
tvansverse. The motion of the wave-particle may,
4 in fact, be likened to the motion of the bob of a
" pendulum, the direction of travel of the wave being

vertically up or down. This form of wave motion

Amaro A is represented diagrammatically in fig. iv. The
ندرا پک‎ gE pd ped figures opposite the dots having the same signi-
plete undulation of a distortional ficance asin fig. ii, the amplitude is the distance
PANE oto 4, and the period is the time taken to travel
through the complete cycle. As before, there are two principal phases of the move-
ment, in two opposite directions, and these are divided each into two semiphases
according as it is towards or away from the normal position of the particle. The
wave length is the distance between two points where the wave-particle is at the
same extremity of its displacement at the same instant, these extremes corres-
ponding to the crests and trough of a gravitational wave,

. Ig. Elastic surface undulations.—When waves of either of the last two
kinds after travelling through the interior of a body reach its surface, they might
produce disturbances of form which would simulate the surface undulations of

008

1509

EI

18 e

x OLDHAM: GREAT EARTHQUAKE OF 1897.

gravitational waves. Apart from this there are, however, as has been shown by
Lord Rayleigh, true surface undulations, closely simulating gravitational waves
but propagated elastically. In these the path of the wave+particle is an ellipse `
whose major axis is vertical and from one and a half to twicethe minor. The hori-
zontal displacement extends but a short way into the mass, disappearing at from
one-eighth to one-fifth of the wave-length from the surface. Below this there is only
a vertical movement. It does not appear that the existence of waves of this nature
has ever been demonstrated, and they have certainly not as yet been separated
in the complicated disturbances of an earthquake, but the possibility of their
existence may be taken as demonstrated, and time and the collection of fuller
details will doubtless lead to their recognition,

20. It may be noted here that whether the surface undulations noticed in great
earthquakes are of the nature of Lord Rayleigh’s waves or not, there can be no
doubt of their existence. They have been observed too often for this to be doubted,
and ample proof of their reality will be found in the pages of the foliowing report.
It must, however, also be noticed that there are apparently two distinct types.
Firstly, there is the long, low, quick travelling wave, which first begins to be dis-
tinctly noticed in the area where the earthquake has ceased to be destructive,
or where the destruction wrought has become much less than in the epicentral
tract. These waves cannot be seen; they are recognised only by a successive
tilting in one direction and then in the opposite, and their rate of travel, “190 to
120 miles a minute, is that of the earthquake shock. Secondly, there are the
much shorter and steeper sided waves which can be seen and recognised as such ;
these may vary in length from 20 feet to 200 yards and in height from a few
inches to 2 or 3 feet, and travel at a rate of 3 to 8 miles an hour.

21. Whatever may be the nature of the laws which govern the movement of
the wave-particle in these surface undulations it is hardly possible that they can
be the same in the case of waves which differ so much in size and rate of move-
ment as the two forms of surface wa ve referred to in the preceding paragraph
This conclusion appears to be supported by the absence of any passage from
the one to the other, but on this point further observation is required.

22. Complication of waves.—lf the earth were a perfectly homogeneous globe,
then, supposing elastic waves of one of the three classes were set up by a dis.
turbance at any part of it, they would be propagated without change. If the
waves of all three classes were set up, they would separate as they travelled, the
waves of elastic compression, travelling at the greatest speed, would outrace the
others ; then at a gradually increasing distance behind would come the waves of
elastic distortion, while lagging more and more behind would come the elastic
surface waves, which not only travel slowest, but have to travel round the surface
of the globe, while the others can take a short cut through it.

23. In the case of the disturbances which are registered by sufficiently deli-
cate instruments at a long distance from the place they start from, and have
consequently travelled through the comparatively homogeneous central core of
the earth, something like this has been observed. In the case too of artificial
disturbances, such as are produced by explosion of gun-cotton, something of the
kind has heen observed where the disturbances have been measured at compara-
tively small distances and have travelled through fairly homogeneous material.

TAR Ld

PREFACE, 21

24. In the case of earthquake waves in the region where the shock is severe
enough to be felt, nothing of the sort has been recognised and nothing of the sort
is to be expected, seeing how heterogeneous are the materials of which the outer
crust ot the earth is composed.

25. The rate at which elastic waves are transmitted depends only partly on
the elasticity, whether that measured by the bulk modulus, or the rigidity ; the
density of the materials is also a controlling factor. A wave may, consequently,
preserve the same rate of propagation, even when it passes from one medium to
another of different density and elasticity, provided that the changes are in exactly
such proportion that the new density combined with the new elasticity give the
same rate of propagation as before. Usually, however, there will be a change in
the rate of propagation ; and when a wave of either elastic compression or distor-
tion passes from one medium into another, in which it has a different rate of pro-
pagation, not only is its direction of travel changed, in other words, not only
does it undergo refraction, but it is also split up and partially converted into
elastic wave motion of the other kind.

26. Seeing how very heterogeneous are the materials of which the earth’s crust
is composed and how frequently waves travelling through it must pass from one
medium to another in which they have a different rate of travel, it is easy to see
that, even if the motion in an earthquake wave was simple to start with, it would
soon become extremely complicated.

11 —Definition of seismological terms.

Having treated of the various possible forms of wave motion, we are ready
to go on to a consideration of the observed facts of earthquake movements, but
before doing so it will be necessary to explain the meaning of certain special
terms which will ke frequently used.

C > BA A ic] + B y

Fig. v. Diagiam in explanation of seismological terms.

27. Whatever be the cause of an earthquake, the disturbance starts from some
place within the crust of the earth. This may be so small that it can be consi-
dered as a point, or more usually have considerable extent. Whatever its ¿size or
form, itis known as the centrum or centre, or, if regarded from another point of
view, as the focus of the earthquake. For convenience we will assume it to be a
point, F, in fig. v, which represents, diagrammatically, a vertical section through the

xil OLDHAM: GREAT EARTHQUAKE OF 1897.

crust of the earth, S S being the surface. Then a straight line drawn vertically from
F to E isthe seismic vertical, and the point E is the epicentre. Where the centre,
instead of being so small that it may be considered a point, is of considerable size,
then the epicentre increases in size to a corresponding extent, and we have an
epicentral or epifocal area or tract.

28. From the centre ۶ the earthquake waves radiate in every direction. To
the epicentre they travel straight upwards ; away from it they reach the surface at
a less and less angle as we get further away. Let F A represent one of these
wave paths : then the angle E F A is the angle of emergence.

29. The front of the disturbance travels at a definite rate, and if the rate of
propagation is equal in every direction it will at any given moment have reached
the same distance from the centre in every direction. The circles drawn in fig. v
may be taken to represent the wave front at successive equal intervals of time
and are known as homoseists. They are surfaces, which appear as lines where
they are cut by the ground level, and if we had a perfectly homogeneous earth
and a perfectly level ground, they would appear as circles. In practice neither
condition holds good and the homoseists or homoseismic lines are irregular in
form.

30. As the earthquake radiates from the centre it becomes less and less violent,
and it is possible to trace lines of equal violence of shock which are known as
isoseists or isosetsmic lines. If a condition of perfect homogeneity prevailed in the
crust of the earth they would coincide with the homoseists; but as the violence
of the shock is much more influenced by the nature of the surface deposits than
the rate of propagation, they are in practice found to pursue an even more
devious and irregular course.

31. The actual violence of the shock doubtless decreases outwards from the
epicentre, but its destructiveness does not necessarily do so. At the epicentre the
thrust, if we are dealing with a wave of elastic compression, is straight upwards ;
there is no sidelong push, and consequently no tendency to overturn walls. As
we recede from the epicentre, and the angle of emergence decreases, the sidelong
motion increases, and there is an increasing tendency to overturn walls and
buildings; the violence of the earthquake is, however, diminishing, and there is a
point where the increasing value of the one and the decreasing value of the other
lead to a maximum of destructive effect. The line along which this takes place
is the meizoseismic line or circle; according to the view we take of the rate at
which the violence of an earthquake diminishes the angle of emergence to which
the meizoseismic circle corresponds becomes 45° or 54” 44. Theoretically, then, the
depth of the focus from the surface should be equal to, or rather over, half the
diameter of the meizoseismic circle.

32. Within the meizoseismic circle the variation in destructiveness of the
earthquake is found to be much less than should be the case if the movement
were purely that of a wave of elastic compression. This is in accordance with
what has been said regarding the impossibility of one form of wave motion
reaching the surface unaccompanied by others, In practice it is. generally found
that there is an area over which there are comparatively small variations in the
destructive force of the earthquake. This lies within the meizoseismic circle, which
is not always recognisable, while outside it there is a much more rapid decrease of
destruction. This inner area is known as the pleistoseismic area.

PREFACE. xili

34. The greater destruction in the pleistoseismic area, as well as the com-
paratively smaller variations in the amount of destruction, are probably due to
a radical difference in the nature of the motion of the ground. Above the focus it
is probable that the ground is lifted bodily and that, besides the molecular motion
of the elastic waves, there is a certain amount of molar motion of the whole of the
rocks lying between the focus and the surface. This would not be confined to
the region vertically above the focus, but would extend outwards ; whatthe exact
limit may be is not known, but it is probable that an angle of 45? from the verti-
cal would include all that was so affected. - For this mass of rock the term core
of the earthquake has been suggested, and it affordsa convenient, and apparently
true, distinction between the central core on the one hand, in which both molar and
molecular movement takes place, and the area or region of propagation on the
other, in which we have only elastic waves, set up by the earthquake and travelling
outwards with a decreasing energy and violence.

34. The term seismic area was applied by Mallet to the region over which the
earthquake was felt; of late years, when the improvement of instruments for
recording earthquake movements has rendered it possible to detect the disturbance
caused by a great earthquake atlong distances, beyond the utmost limit at
which it can be recognised without the assistance of these instruments, theterm
microseismic has been applied to these unfelt movements and the seismic area of
Mallet has sometimes been distinguished as macroseismic. The terms seem incon”
venient and hardly appropriate, for some of the waves in the area where the
disturbance would be described as microseismic are larger in size than any of
those in the macroseismic area. It seems consequently best to retain the terms
seismic area in its original sense and to describe the movements that take place
outside it as cryptosetsmic reserving microseismic for those minute earthquakes
which nowhere become sensible to unassisted observation.

35. Two more terms may be explained with advantage, though they are not
specially seismological, and these are the words resultant and component. If a body
is acted on simultaneously by two distinct forces, the final result will be the same
as if each force acted singly and successively. Where the two forces do not act in
the same or opposite directions, this leads to the construction known as the parallelo-
gram of forces; in fig. vi let O represent a body acted on by two forces in the

directions A and B, andlet OA and OB
respectively represent the magnitude
of the forces, that is, the distance
through each singly would displace
O in, say, one second of time ; com-
plete the parallelogram O A R B,
A then R represents the position which
O will occupy at the end of a second,
and OR will represent the direction
and distance that O will travel in that time. In other words O R represents
the direction and magnitude of a single force, called the vesultant, which
will produce the same result as the two separate and simultaneous forces O A and
O B. Similarly, if we have a single force O R, it may be regarded as composed of
two separate forces O A and O B which are called the components. Moreover, as

-A

Fig. vi.

XIV OLDHAM: GREAT EARTHQUAKE OF 1897.

the distance to which a body, free to move, is displaced in a given time is a measure
of the force affecting it, the same words are applied in a similar way to motion,
which may be compounded into a single vesultant, or resolved into one or more
components.

The components are usually, but not necessarily, considered to be at right
angles to each other, this being purely a matter of conveniénce or convention.

36. Though possibly self-explanatory it may be well to point out the sense in
which the words wave front and wave path are used. The wave front is the same
as a homoseist ; it is the limit to which, at a given moment, the wave motion has:
reached. Ina perfectty homogeneous medium this is everywhere at the same
distance from the focus, but if the medium is like the material of which the earth
is composed, such that the wave motion is transmitted more rapidly in some
directions than in others, the distance of the wave front from the focus will vary
from place to place. The wave path is a line along which the wave motion may be
supposed to travel. There are an infinite number of such lines radiating from
the focus, and each cuts the successive wave fronts always at right angles. From
this it follows that if the wave motion is propagated with equal rapidity in every
direction the wave paths will be straight lines, but if the wave motion is propagated
more rapidly in some directions than in others, the wave paths will bend round, so
that they shall always form a right angle with the varying direction of the wave:

front.

III —Nature of earthquake motion.

From the earliest times the fact that there are distinct types of earthquake
motion has been recognised in Italy. Earthquakes are there divided into four
classes, orizzontale or horizontal, in which only a to and fro motion is felt ; oxduloso:
or undulating, in which the ground appears to be thrown into waves and the
sensation is like that of a ship rocking on a gentle sea ; sussultorio or palpitating,
in which there is a distinct vertical or up and down motion ; and vorticoso or vorti-
cose. The justification of this last distinction, which is based on the effect of the.
earthquake, has been questioned but it appears sound ; the subject is discussed at
length in Chapter XIV of the following report.

38. Though these distinctions, which are popular rather than scientific, have
long been recognised, it is only comparatively recently that instruments have been
devised by which the actual movement ofthe ground has been traced. Of the
exact nature of the movement within the core of a really destructive earthquake, or
even of that same earthquake outside the core so long as it retains a high degree
of destructive power, nothing detailed is yet known. In these circumstances the
instruments and their records are alike lost in the ruins of the building which

contained them. j :
39. Beyond this region of extreme destruction, but where the shock might still

be described as destructive, records have been obtained. It is found that, the.
earthquake sets in with minute rapid preliminary tremors which increase in
amplitude and period tillthey suddenly give way to one or two very much
larger movements, and these again are succeeded by a series of waves of longer:

and longer period.

PREFACE. XV

40. Leaving, for the moment, the consideration of the preliminary tremors
and of the long slow afterwaves, and confining attention to the central parts of
the disturbance, which constitutes the sensible shock, it seems necessary to dis-
tinguish the character of the movement in small and great shocks. In those
which nowhere reach a degree of violence sufficient to cause alarm, and much less
any actual damage, it is probable enough that the motion is largely due to
elastic vibration propagated directly from the focus to the surface through the
substance of the earth. In the case of severe earthquakes, this form of motion
is also present, but overpowered by waves of a different character, which are prob-
ably analogous to the elastic surface undulations of Lord Rayleigh, and originate
at the surface, within the epifocal area, whence they spread outwards in every
direction.

41. A peculiarity of this sensible earthquake motion is that, as the disturbance
radiates, its duration at first increases and then appears to diminish. This was
very noticeable in the case of the earthquake of 1897 ; within the epicentre the dura-
tion of the great shock did not exceed two minutes ; at Calcutta it was certainly
5 or 6, and the maximum duration seems to have been at about 500 miles from
the centre, reaching some seven minutes. From hefe the reported duration
diminishes till at the furthest limit, in Ahmedabad, it is reported to have been but
a second or two. This does not, however, mean that the disturbance of the
ground only lasted for that period, but that, except fora few seconds, the move-
ment was too slow and gentle to be felt.

42. In the area of propagation of an earthquake, this movement, which can be
felt, and to which the damage done is due, is quite superficial. It has been found
by experiment in Japan thatan excavation of only 20 feet in depth is enough to
reduce the wave motion of an earth juake to a very small fraction of what it is
at the surface. Another fact, pointing to the same conclusion, is the rarity
with which an earthquake is felt underground in mines, even where these are
shallow. Inthe rare cases where an earthquake has been felt underground, like
that referred to in Appendix H tothe following report, it is probable that the
mine lay within the core Of the earthquake, and that the motion felt was due to
those molar displacements of the ground which have been referred to in para. 33-
The 1897 earthquake was not felt underground in the mines of the Bengal
coalfields ; it was felt and caused some damage in the coal mines at Makum in
Upper Assam; but these mines are really over ground, being driven into the
side of a hill and not descending below the general surface level, the conditions
are, therefore, not strictly comparable to those of a mine sunk in open and
tolerably level country,

43. If the felt earthquake wave is of the nature of Lord Rayleigh’s elastic
surface undulations, the restriction of the shock to the surface is in accordance
with the theory, for at a short distance the range of motion would become too
small to be felt below the surface, It may also be explained, in the case of purely
elastic condensational or distortional waves, by the greater play which is possible
at the surface than in the body of the earth, While the wave is being transmitted
through the rocks which constitute the crust of the earth, the movement of '
each wave-particle is restrained by the resistance of those in front, to which it
has to communicate its motion ; but when the wave reaches the surface there is

xvi OLDHAM: GREAT EARTHQUAKE OF 1897.

nothing in front to restrain the movement of the wave-particle and it can cone
sequently range further out from its normal position. The effect of this greater
freedom and range Of motion where the wave comes out to a free surface is very
marked, and will be referred to in the chapters dealing with earth fissures and
landslips.

44. To this greater freedom of motion at and near the surface of the ground we
must, at any rate, to a large extent, attribute the more noticeable and destructive
character of the shock at the surface of the ground. Of the nature of the move-
ment, all that can be said here is that it is characterised by an extreme com-
plication and intricacy of the path of the wave-particle. At one and the same time
it is subjected to wave motion of varying amount and period, and of each of the
three distinct types described in paras. 9 to 21.

45. In spite, however, of this complexity, a prevailing direction can often be
recognised, which may or may not change during the period for which the earth-
quake lasts, Asa rule there is no relation between the direction of this movement
and the direction in which the epicentre lies, that is, from which the shock is
traveling. In thecase of very severe earthquakes it may happen that many
of the buildings and monuments overthrown point in a direction towards, or away
from, the epicentre. Where this is so they may be used for determining the
position of the focus, but the method, as referred to in Appendix D, requires both
Care and skill in its application.

46. So far reference has only been made to what can be observed without
instrumentalaid. Ifthe same disturbance is studied with the aid of seismographs
it is found that the duration is about double of what can be felt, and that it
follows the samelaw. ‘That is to say, it first increases as the epicentre is receded
from, and then decreases, only a smaller and smaller fraction of the whole
disturbance being of a character which can have any effect on the instrument.
What takes place may in fact be likened to the changes that take place in the band
of ripples set up ina pond by astone falling into it. As they spread outward
the width, not only of each individual wave, but of the whole band, increases, and
then, if the pond is large enough, one wave after another may be seen to grow
so long and flat that it ceases to be discernible, and finally the pond resumes its
originally undisturbed condition.

47. This spreading out of the waves is very conspicuous in the records ob-
tained by the delicate instruments specially designed to detect the movement
which takes place at distances beyond where the earthquake ceases to be re-
corded by ordinary seismoscopes. In these the preliminary tremors which have a
duration to be measured in seconds withir the seismic area, outrun the larger waves
by periods which may range up to half an hour. They are probably due to
elastic vibrations propagated through the body of the earth, and in the case of
the earthquake of 1897 it was possible to recognise that they present two phases
which may be interpreted as being due to thearrival of the condensational, and
of the slower travelling distortional, waves respectively.

48. This division of the preliminary tremors into two phases had not been
definitely recognised before 1897, but it seems probable that it may be found in
the records of other earthquakes when they come to be re-examined in the light
of the distinction recognised in the case of the 1897 earthquake. It lends such

PREFACE, xvii

Support to the hypothesis that the preliminary tremors are transmitted through
the earth, as almost amounts to proof. Were they transmitted through the super-
ficial crust of the earth, a separation of the two classes of waves would be im-
possible for the reasons given in para. 25, but, in the more homogeneous matter
constituting the central core of the globe, such a separation is what would inevit-
ably happen, through the more rapidly travelling condensational waves outracing
the slower travelling distortional. Moreover, itis only at great distances, where
the portion of the wave path which lies within the surface crust, or,say, the outermost
30 miles, of the earth, forms but a small fraction of the total wave path, that such
separation could take place. If, as seems probable, this hypothesis stands the
test of further experience, it is certain that in the majority of cases the preliminary
tremors of distant earthquakes, as recorded by the instruments in present use,
are those due to the distortional waves, those due to the condensational being
ordinarily too feeble to be recorded.

49. Following on these preliminary tremors come the great surface waves
so well shown in the diagram on Plate XLI. which have expanded from a length
of 2to 3 miles and a period of 2 or 3 seconds to a length of 30 miles and a period of
25 seconds, while their duration, instead of being a few minutes, ranged over half
an hour, or, if the slighter and less distinctly recorded waves are included, of
several hours.

IV.—The elements of earthquake motion which can be measured.

There are certain elements of the motion of an earthquake which can be
measured and expressed in figures. These are: (1)the vate of transmission or
velocity with which the wave motion travels from place to place, (2) the amplitude-
(3) the period, (4) the velocity of movement of the wave-particle, and (5) the acces
leration of the wave-particle.

50. The vate of transmission is readily determined if we have accurate records
of thé time of arrival of the earthquake at different places, whose distances apart
are known. Theoretically, if the time of arrival at three places is known, it should
be possible, provided the times are not simultaneous, to determine the rate and
direction of travel of the earthquake wave. 18 practice, however, it is not possible
to obtain records of sufficient accuracy, and this apart from any fault of
the observations orrecords. The wave motion is not a simple one, and to
compare the times it would be necessary to be certain that they referred to one
and the same phase of the movement, From what has been said of the constant
variation of the character of the wave motion from place to place, it will be seen
that this is impossible ; consequently, it is only possible, in practice, to obtain a fairly
accurate estimate by combining a large number of records from many different
places.

51. Another element of uncertainty in all determinations of the rate of trans
mission of the earthquake waves lies in the fact that all we can obtain by direct
observation is the apparent rate of travel along the surface of the earth, which
may be very different from the true rate of travel through the interior of the earth.
In fig. v, for instance, if the successive circles represent homoseists at intervals of
one second, then the true distance travelled by the wave in the first second after

xviii OLDHAM: GREAT EARTHQUAKE OF 1897.

the earthquake reaches the surface is represented by A'A while the distance it
would appear to have travelled along the surface is represented by E À;
similarly, B'B and A B represent the ¿rue and apparent distances travelled in the
next second, and so on.

52. It will be seen from this that not only will the apparent rate of travel along
the surface be different from the true rate of travel through the earth, but that it
will vary as the distance from the epicentre varies. This will be true of the waves
which are transmitted through the earth whether the wave paths be straight lines,
as represented in fig. v, or curved, as is more probable. In the case of those
great earthquakes which have been exhaustively examined, no such variation has
been recognised ; on the contrary, the rate of transmission seems to be practically
uniform, and it is this which has led to the conclusion that, in the case of great
earthquakes at least, the movement which can be felt is due to waves propa-
gated along the surface, and that in their case the apparent and true velocities
are identical. In the case of the preliminary tremors, on the other hand, the
apparent rate of transmission varies largely accordinp to the distance from the
epicentre, and this leads to the conclusion that they are transmitted through the
earth, and that in their case the true and apparent velocities are not the same.

53. The amplitude of the wave has already been explained (paras. 6, 13, 18) و‎
it is usually expressed ia millimetres or in inches ; double the amplitude is the
range of motion of the wave-particle, or more correctly would be, if the motion
was a simple harmonic one. The error introduced in practice bv assuming the
amplitude to be half the range of motion is, however, in all probability immaterial.

54. The period of the wave has also been explained, and need not be further
treated. It is usually expressed in seconds of time.

55. The velocity of movement of the wave-particle is a quantity which varies
from moment to moment. Each particle asit starts from a position of rest com-
mences moving slowly, then moves more and more rapidly, till it attains its
maximum velocity, after which this decreases till the particle once more comes

to rest before commencing its return movement, in which the same succession of
a gradual increase of the rate of movement up to a maximum and a sub-
sequent gradual decrease takes place. The velocity of movement at any moment
is expressed in feet, inches, or millimetres per second, and does not mean that the
particle actually travels so many feet in one second, but that, if it wereto continue
moving at the same rate for a whole second, it would then have travelled the
given number of feet, inches, or millimetres.

56. As a rule, we are only concerned with the maximum velocity cf the wave-
particle, that is to say, the most rapid rate of movement attained ; like all the lesser
velocities of wave-particle it is only momentarily attained, but is the only one to
which a numerical value can be given which is independent of time. The lesser
values vary from moment to movement and from zero to the maximum ; but it is
obvious that there can be but one maximum value, and that this value can be given,
and remains true, apart from any statement of the exact instant at which it was
attained.

57. The acceleration, or rate of change of velocity, of the wave-particle is one
of the most important elements to be determined, for on this depends the violence
and destructiveness of the earthquake. As explained, in para. 55, the velocity

PREFACE. XIX

of the wave-particle is never the same for two successive instants: it is constantly
changing, but the rate of change is not constant. Like the velocity, it is nothing
at one point of the path of the wave-particle, increases to a maximum, and
then decreases again, being zero when the wave-particle is at rest or at its maxi-
mum velocity, and attaining a maximum four times in each complete wave period,
at positions half way between the maximum: and zero velocity, which are reached
twice only in each complete undulation. |

58. As it is important that there should be a clear understanding of what is
meant by acceleration, a brief explanation may not be amiss. Suppose a body,
such as a railway train, is standing still, and that, by the action of an external
force, such as a locomotive, it is set in movement, and that at the end of the first
second it is moving atthe rate of one foot per second, at the end of the next second
atthe rate of two feet per second, at the end of the third at the rate of three feet per
second ; then the acccleration of that train is one foot per second per second.
That is to say, at the end of each second its velocity is one foot per second greater
than at the beginning.

59. The instance taken is a very simple one, where the acceleration is supposed
to remain constant for at least three seconds ; in the case of a wave-particle, on the
other hand, it does not remain constant for two successive instants. Just, however,
asit is possible to define the velocity at any instant, so the acceleration at any
instant may be defined as being so many feet, inches or millimetres per second per
second; by which itis not meant that the given rate of acceleration is maintained
for a whole second, or for any finite fraction of a second, but that, if the actual
rate of change of velocity at the given instant were maintained for a whole second,
the velocity would at the end of it be so many feet, inches or millimetres per second
greater than at the commencement.

60. It isimportant that the distinction between the velocity or rate of move-
ment, and the acceleratíon or rate of. change of velocity, should be clearly under-
stood. It is possible to havea high velocity combined with a low acceleration, or
a low velocity with a high acceleration, the two being quite independent. It may
also be necessary to state that the word acceleration is, in this connection, equally
applied to what would ordinarily be regarded as retardation : a body which,
moving atthe rate of ro feet per second, slows down at the rate of 5 feet per
second per second is said to have that acceleration equally with one that increases
its speed at that rate. The word is in fact used as equivalent to rate of change
of velocity, and irrespective of whether the change is one of increase or decrease.

61. Asin the case of the velocity of wave-particle, itis only the maximum
acceleration whose determination is ordinarily necessary or attempted.

62. The intensity of an earthquake.—Besides the elements above mentioned,
to which numerical values can be given, attempts have been made to define in a
similar manner a vague and uncertain quantity called the intensity of an earth-
quake. The bestknown and most often used of these so-called scales of inten-
sity is that known as the Rossi-Forel, which, because itis so frequently used and
because in the absence of anything better it may be better than nothing, is given
below, Of it, all that may be said at present is that, though to a certain extent
it may serve for the comparison of different earthquakes, each being taken at its
maximum intensity, it does not serve as a scale for the gradual dying out of a

xx OLDHAM: GREAT EARTHQUAKE OF 1897.

severe earthquake. For instance, degree IV specifies, as a part of its characters,
the disturbance of doors, etc.; in the case of the 1897 shock, this was recorded at
Sehore in Central India and at Bangalore, yet these places were beyond the
region where the earthquake could be distinctly felt, and would fall, at the outside,
in degree I of the Rossi-Forel Scale. Another objection is its limited range. The
area which, in 1897, was covered by the maximum degree, No. X, of the
Rossi-Forel Scale is as large as England, and within it there were variations of
violence, or, as ordinarily called, intensity, which would require the scale to be
extended for at least another 5 to 10 degrees.*

63. The fact is that the word intensity as ordinarily used, and as itis attempted
to be defined in these scales—which were intended to serve in seismology the
purpose which Moh’s scale of hardness serves in mineralogy—is altogether too
vague and indefinite a term to be used in scientific description. Prof. Menden-
hall has pointed out that there are three distinct quantities which have been or may
be used as measures of the so-called intensity at a given locality. They are—

(1) the maximum velocity of wave-particle,

(2) the maximum acceleration of wave-particle,

(3) the energy of the earthquake, which may be measured by (a) the energy
contained in a unit volume of rock set in motion by the wave, (4) the
rate at which energy is transmitted across a unit area of the wave
front.

Of these, (3) in one of its forms is the most rational and scientific equivalent to
be represented by the word intensity, and, of the two forms, (b) is that which Lord
Rayleigh has adopted in his * Theory of Sound” as the measure of intensity of

radiation.

64. Unfortunately the intensity, in this sense, is incapable of direct measurement,
and, so far, has not been satisfactorily determined in the case of any earthquake.
As ordinarily used, the word is applied to the greater or less impression produced
by the earthquake on the senses or the greater or less destruction wrought by it.

1 The Rossi-Forel Scale of Intensity.—Translated by Dr. C. Davison.

I. Recorded by a single seismograph, or by some seismograpbs, of the same model, but
not by several seismographs of different kinds; the shock felt by an experienced
observer,

II. Recorded by seismographs of different kinds; felt by small number of persons at rest.
111. Felt by several persons at rest; strong enough for the duration or the direction to be
appreciable. i
IV. Felt by persons in motion; disturbance of moveable objects, door, windows, cracking
of ceilings.
V. Felt generally by every one; disturbance of furniture and beds, ringing of some bells,
VI. General awakening of those asleep ; general ringing of bells, oscillation of chandeliers,
stopping of clocks; visible disturbance of trees and shrubs. Some startled persons
leave their dwellings,
VII. Overthrow of moveable objects, fall of plaster, ringing of church bells, general panic,
without damage to buildings.
VIIL Fall of chimneys, cracks in the walls of buildings,
IX. Partial or total destruction of some buildings.

X. Great disasters, ruins, disturbance of strata, fissures in the earth’s crust, rock-falls

from mountains.

PREFACE. xxl

This is now recognised to be due to variations in the rate of acceleration of the
wave-particle, and this might be used as a measure of what has been called
the intensity, but that the degrees of the scale depend to some extent on the
destructiveness or destruction produced by the earthquake, while the accelera-
tion is more properly a measure of what Mendenhall has called the destructivity
or power to destroy. The word proposed is, however, an awkward one, and the
simpler word violence seems more suitable ; of this the rate of acceleration of wave-
particle is a measure, and in proportion as this increases the earthquake
becomes more distinctly sensible and more and more destructive. Yet the
destruction actually caused must not be taken as a measure of the violence
without observing proper precautions, for it is determined both by the power of
the earthquake to destroy and the power of the object to resist destruction. .
This is equally true whether we have to deal with buildings overthrown, landslips
from hills, or fissuring of the ground, and consequently, in attempting to compare
the violence of the earthquake at two different places, we must consider not only
the effects actually produced, but the particular local circumstances governing
their production.

65. Bearing these considerations in mind the word violence may well be used
instead of acceleration in all those cases where thereis no possibility or desire of
giving it a precise numerical value. The word will, in fact, be used in the follow-
ing report as substitute for intensity, the latter word being reserved till the time
when, as the means of seismological observation improve, it can be used in its
more proper sense.

V.—The means of measuring earthquakes.

66. The most satisfactory and trustworthy means of obtaining the values of
the elements of earthquake motion which are capable of measurement is that of
instruments specially designed for the purpose. These fall into three classes
according to their purpose; seismoscopes, which merely record the fact of an earth.
quake, of sufficient severity to affect the instrument, having occurred; seise
mometers, which are intended to measure one or more of the elements of the
earthquake motion; and setsmographs, which record in greater or less detail
the actual course of that motion from the commencement to the close of the
shock.

67. In principle the instruments may be classed according as they consist of
(1) a mass which is set in motion by the earthquake ; (2) a mass which is so
supported as to remain more or less steady, while the earth moves underneath it ;
and (3) an arrangement by which the actual tilting of the surface is measured.
Two or more of these may be combined in the same instrument, and on one or
more of these principles a vast number of instruments of varying design have
been constructed or proposed. To attempt a description of them here would
occupy too much space, and for an account of them and the capacities of the
various types reference should be made to Prof. Milne’s two books.

68. Apart from special instruments, much information may be obtained from
' observations made at the time of the earthquakes, or from the effect produced by
it; in the case of very violent shocks this is the principal source of information,

xxii OLDHAM: GREAT EARTHQUAKE OF 1897.

69. The vate of transmission is obtained from observations of the time at which
it was felt, and it is necessary that these should be accurate at least within one
minute of time, even in the case of earthquakes which are felt over a large area.
In the case of lesser earthquakes an even greater degree of accuracy is required
if the results are to be of any value, yet even here, if the observations are
sufficiently numerous, a much greater degree of accuracy in the result can
be obtained tham the errors of each individual observation would appear to
justify.

70. Another necessary precaution is to compare similar phases of the shocks,
and in practice the only phase which can be identified, with even an approximate
degree of accuracy at different places, is the commencement. When a sufficient
number of trustworthy observations are available, they may be treated either by
means of the mathematical process known as the method of least squares, or in
the simpler method adopted in Chapter III, which will usually give results as
accurate, though less precise, than those obtaíned by the more elaborate method.

71. The replies to circulars issued after the earthquake of 1897 showed that the
stoppage cf clocks is popularly supposed to be a particularly accurate method of
recording the time of an earthquake. This, however, is not the case; not only is
there no means of determining the particular phase of the shock to which the
stoppage was due, but in some cases theclock may not stop till some time after
the earthquake has ceased. One authentic case is on record—the earthquake of
24th June 1870 at Athens— where the pendulum of an astronomical clock was
observed to be knocked against the clock case by the earthquake, yet the clock
only stopped some five minutes after the earthquake had ceased.

72. Theamplitude of the wave and the velocity and acceleration of the wave-
particle can be determined, in the case of earthquakes of sufficient violence to
cause damage, by observation of its effects. The methods and formula in use
are sufficiently explained in Chapter V and in Appendix C.

VI.—kEarthquake effects. `

73. Cracks in the walls of buildings are probably in large part due, and in
lesser degree fissures in the earth, to the bending of the surface, as it is thrown
into undulations by the passage of the earthquake wave. With this exception,
all the effects of an earthquake, whether they take the form of visible damage or of
the sensations produced at the time, are entirely due to inertia.

74. Everything tends to remain in the same condition of rest or motion in which
it is at any moment, and is only by the action of an external force that a change
can be produced. A pillar, standing upright, has no tendency to fall over,
but when its base is suddenly displaced, as by the motion of the ground in an
earthquake, the mass of the pillar tends to maintain its original position and a
strain is set up at the base. If this strain is great enough the pillar will be tilted
up onthe edge away from the direction in which the base was moved, and if the
movement be great enough the pillar will topple over.

75. lf, from the form of the body affected, it is not free to fall or bend in
the direction frcm which the motion starts, the velocity of movement of the wave-

PREFACE. gro

particle will be communicated to it, and, when the next semiphase of wave motion
sets in, its 7zertza, or, as commonly expressed, momentum, will tend to make it
continue to move with the direction and velocity which have been imparted to it.
A strain will consequently be set up, just as in the case considered in the last
paragraph, and in this way not only may pillars be overthrown but the ends of
walls may be shot off, or the soil resting on a hill side, or even the face of a cliff, be
dislodged, as described in Chapter VII.

76. The magnitude of the strain set up is governed, not by the velocity, but by
the acceleration of the wave-particle. This is easily illustrated experimentally.
Let a match-box, preferably full, be stood on end on a sheet of paper resting on a
table ; if one edge of the paper is taken hold of, the paper and match-box can be
moved over thetable and,if the speed of movement be only gradually increased,

. it will be found that the motion may become quite rapid without causing the

match-box to overturn ; if, onthe other hand, the sheet of paper be put suddenly
in motion, the match-box will fall over backwards, or if after having been put in
rapid motion it is suddenly stopped, it will fall overforwards. Similarly, if a
train is started or stopped gently it is very difficult for a passenger to recognise
the moment of starting or stoppage, but if an unskilful, or careless, driver is in
charge of the engine, the stoppage, or starting, may be a jerk that is very notice-
able. In both these instances it is not the rate of movement, but the rapidity of
change of rate of movement, that is to say, the acceleration, which produces the
effect. Similarly, in the case of an earthquake the sensation produced, or the
damage done, is due to the greater or less suddenness of the movement and not
to the actual velocity attained.

77. There is another way in which solid bodies may be affected by the earth-
quake wave, which depends, not on the acceleration, but on the velocity of the
wave-particle, If, instead of a tall pillar, we take the case of a flat stone lying on
the ground, it is obvious that any overturning is impossible. The motion of the
wave-particle will be more or less completely imparted to the stone, and when the
next semiphase of the wave motion sets in, the inertia or momentum of the
stone will tend to make it continue moving with the direction and velocity which
has been imparted to it. If the momentum imparted be more than sufficient to
overcome the resistance of the attachment of the stone to its support, it will be
projected through the air, and the distance to which it is projected is dependent on
the direction and velocity with which it started on its way. Yet, even here,
acceleration is not without its influence, for, where a body is projected upwards,
whether vertically or in a slanting direction, the acceleration of the wave-particle
must at least exceed that due to the action of gravitation, or the attraction of the
earth on a free falling body near its surface.

78. This statement may easily be illustrated by placing a smallstone in the
upturned palm of an open hand. Ifthe hand is raised and then lowered slowly
the stone will not leave it, butif the movement is made quicker and more quicklv
a limit will be reached when the stone leaves the hand and rises in the air; this
means that the rate at which the stone increases its velocity of fallis less than

1 This varies slightly from place to place, but may be taken as about 32 ft. per second per
second,

Fr»

xxiv OLDHAM: GREAT EARTHQUAKE OF 1897.

that at which the hand increases its rate of downward movement. Once, however,
that this limit is passed the distance to which the stone travels through the air is
governed solely by its initial velocity and direction of movement, and if any two of
these three is known the other can be calculated.

79. Though a consideration of the methods of construction, which should be
adopted to prevent or lessen the damage caused by earthquakes, would be foreign
tothe purpose of this introduction, it may be pointed out that the foregoing
paragraphs show that one of two principles should be used as a guiding one,
either great strength or greatlightness.: Seeing that the strains set up are
dependent on two factors, the inertia of the object affected and the acceleration
of the wave-particle, we must, as we cannot control the second, either diminish the
first, and so lessen the magnitude of the strains, or increase the strength of the
structure so that it shall be able to withstand the strains set up ; in other words,
and to use Prof. Milne's simile, the ideal should be either the wicker basket or the
steel safe.

VII.—The causes of earthquakes.

80. An earthquake may be produced by any cause, of the nature of a shock
or blow, by which an elastic wave or waves are set up inthe earth. Thefall of a
mountain, or even a house, the explosion of a mine or magazine, or the work-
ing of a large steam hammer, may set up a disturbance which can be feltas a
earthquake within a certain distance of the origin. Excluding these disturbances,
which originate naturally or artificially at or above the surface of the ground,
and restricting the term to those natural disturbances which originate within the
earth, earthquakes have been recognised as being produced by causes which divide
them into three distinct classes, namely, (1) Rockfall, (2) Volcanic, (3) Tectonic.

81. The vockfall earthquakes are due to the falling in of the roofs and sides
of subterranean hollows or caves. They are feeble, of very small extent and may
often be only noticeable as sounds, and not as sensible shocks.

82. The volcanic earthquakes are those due to the activity of volcanoes, with
which may be included earthquakes due to the rending open of fissures by the
sudden development of steam under high pressure. This last cause has been
inferred in the case of some earthquakes, but its reality has never been
proved ; volcanic activity is, however, well known to be associated with earth-
quakes, and these may sometimes be of very great severity, though always local
in their extent. Within a very moderate distance of a town which has been laid
in ruins the shock may be quite insensible.

83. To the class of tectonic earthquakes belongs by far the most preponderating
majority of earthquakes and all those which can be classed as really great, on
account of their violence or extent. They may be regarded as invariably due
to the sudden relief of strain, as opposed to the volcanic earthquakes properly
so called, which may be regarded as due to the sudden development of strain.

84. Leaving on one side all question of ultimate cause, the known facts of
geology show that the rocks of which the crust of the earth is composed have
been subjected to great deformations. In some places they have been tilted, bent
and compressed till they only occupy half their original horizontal extent; in

PREFACE, | XXV

others they have been broken through by fractures and one side raised by hun-
dreds and even thousands of feet with respect to the other. All these deformations
involve the exercise of great forces, which are opposed by the natural rigidity
and cohesion of the rocks; where this resistance is small the deformation may
go on gradually : where it is somewhat greater the strain will accumulate till
the resistance is suddenly overcome and an earthquake is the result. The
greater the power of resistance the greater will be the strain which accumulates
before the final relief takes place, and the greater the resulting earthquake.

85. Tectonic earthquakes have been further classified according to the form
of the disturbance to which they were due into fault, heave, overthrust and fold
earthquakes. Leaving the last out of the question as a very doubtful class, the
separation of the others appear to be of little use and to be rather fanciful than
philosophical. All alike are the result of the sudden relief of accumulated strain,
and the exact form of the fracture to which this is due appears to be an unessential
detail; added to which the different forms of fracture accompanied by dis-
placement to which the terms fault, heave, and overthrust are applied pass into each
other without any defined line of demarcation. The term fault is used when both
the fissure and the movement are more nearly vertical than horizontal ; in a heave
the fissure is also nearer vertical than horizontal, but the movement is more side-
long than up and down; while in an overthrust the plane of the fracture is nearly
horizontal and the horizontal displacement consequently greater than the vertical.

86. It is important to note that the tectonic earthquakes are in no way con.
nected with volcanic action, and the fears so generally felt in Assam and Northern
Bengal that the earthquake of 1897 would be followed by the outburst of an active
volcano were not only ungrounded, but opposed to all that is known of the action
of the natural agencies by which earthquakes are produced.

VIII —Secondary earthquakes.

87. All great earthquakes are accompanied by others which, as they are to a
certain extent the result of,the principal one, may be called secondary earthquakes,
These are of two kinds, the first of which agrees more or less closely with the prin-
cipal shocks as regards the position of the focus, though succeeding it in point of
time, while the shocks of the other kind coincide with the principal one in point of
time, but start from a different focus. A third class might be made out of shocks
which to a certain extent combine the characteristics of the two first mentioned.

88. Aftershocks.—The shocks of the first sort are known as aftershocks;
they succeed the great earthquake, at first with great rapidity, but the intervals
gradually get longer and longer till the average number of earthquakes becomes
normal once more, The cause of the aftershocks is not difficult to understand.
Every great earthquake is accompanied by more or less displacement in the
interior of the earth; this gives rise to a new distribution of strains, and the
gradual relief of’ these new strains causes a succession of earthquakes, each of
which again causes a redistribution of the strains, and the process goes on till
equilibrium is more or less completely restored. |

89. These aftershocks have been carefully studied in Japan, and it has been
found that their decrease in frequency approximates to a definite law, such that, if

xxvi OLDHAM: GREAT EARTHQUAKE OF 1897.

two lines be drawn at right angles to each other, as in fig. vii, and divided

QA جوف‎ AS 070: 00

Fig. vii Curve of frequency of aftershocks.

mto parts of equal length which shall represent equal numbers of earthquakes in
the one and equal periods of time in the other case, and a series of points is plotted
such that their distances from the one line represent the number of shocks felt in
any given period of time—hour, day, week., etc.—and their distances from the
other, the number of such periods that have elapsed since the great shocks, then
the line joining these points will come close to a parabola which becomes asymptotic
with the two lines at right angles to each other.

go. This will be better understood from an inspection of fig. vi, which represents
the curves of frequency of the aftershocks of two Japanese earthquakes, as deduced
by Prof. Omori. It will be seen that the frequency of the aftershocks is great
at first, but the decrease in frequency is alsogreat; as the absolute frequency
diminishes the rate of decrease also diminishes, till, when the excess over the
normal becomes small, the rate of decrease becomes also very small and the curve
becomes almost parallel to the line representing the interval of time.

91. The actual course of the line representing the frequency of the after-
shocks does not, however, follow the curve exactly, for among the aftershocks
there will be some of greater severity than the average, and as each of these
has its own series of aftershocks, superimposed on those of the main shock, there
will be temporary increases of frequency above, compensated by temporary
decreases of frequency below, the general average represented by the smooth
curve. e d

92. Sympathetic earthquakes.—The existence of secondary earthquakes ot the
second classhas never been positively established, but is probable. They are

PREFACE. XXVII

independent shocks which originate from their own centres at or shortly after the
passage of the main shock, and, as the time of their occurence is determined by
that of the greater shock, they may be called sympathetic earthquakes.

93. It is conceivable that a portion of the earth's crust may be in such a state
of strain that a very slight disturbance may be the determining factor in start-
ing an actual rupture. In these circumstances it is possible that the molecular
displacements due to the passage of an earthquake may be sufficient to deter-
mine the final fracture and so give rise to a sympathetic earthquake; that is
to say,to an earthquake which is unconnected with the primary one, so far as
the place aud cause of origin are concerned, but is so far dependent on it that the
moment of its occurrence is decided by the passage of the earthwaves of the
greater shock.

94. There are many recorded cases of earthquakes which, occurring atthe
same time as greater ones, have been regarded as sympathetic, but it must be
confessed that the connection has in no case been proved. As a rule, the coinci-
dence in time is only approximate, and where it has been close enough to permit
the supposition of acausal connection, there always remains the possibility of
_the coincidence being merely accidental. The existence of sympathetic shocks,
though they are referred to and described in text books, must still be looked upon
as conjectural rather than established.

95. Sympathetic aftershocks.—A third category might be made of shocks
which, though they start from a centre which is neither identical nor contiguous
with that of the great shock, yet follow it and are due to the alterations of strains
caused by the displacements which accompanied the principal shock. In the
case of the 1897 earthquake there were, during the autumn months of the year,
a number of small earthquakes in the Darjiling district of the Himalayas, which
evidently originated there, and did not spread into that district from a distant
centre. The Darjiling district lay well outside the epicentre of the principal
shock, yet these shocks, which originated there, and were much more frequent
than the normal number of shocks occurring in that district, must have been
indirectly due to the great shock; that is to say, the displacements of the earth's
crust under the epicentre of the great shock caused a change and increase
of the strains in the Himalayas outside that tract, which resulted in an increased
number of earthquakes. These were not aftershocks in the ordinary sense
of the word, for they did not originate from the same focus or from an extension
of it, while they were equally not svmpathetic earthquakes as defined above, for
they occurred at a very different time ; as they combine part of the characteristics
of these two classes of secondary earthquakes, they may be distinguished as
sympathetic aftershocks.

MIA ای‎ into Ye etm + MM d

LED Ho SA NR‏ ^ رن wa‏ وو nif FE‏ ا

ICE v, ! can € SÉ Ge? v. Së ANN: di^ m SIE ki Gë 7

5۰ ae RM Do MID Ki ا‎ än Zeen EA را‎ VES

y ا : ا‎ A E Sd dn uod SS" Weier zët RER E.
Odd iJ 74 i 07 یہ‎ " T Sea er SO 090

Ei adde A S 00 FERE ran m IN
GREEN ASLA BESA Er ا‎
"RE Meet ۷و و‎
GENTE TS ON FORE ANE uma ا سو‎ med

ELEC AC pd e LS xu loe oeh SEH en)
: او و وک وو و‎ Tae UE EE UN. qe 5 p d
Pd SOM El ا‎ MN SE

m

€ ‘ = TEAS E à Y
= KS Eet e
y ۱ S
à K 2 و و‎ 258 Ke EK end j i bi
í Mnt j pala E me "M Ge ڈیڈ‎ EN de
h 4 1 As PEE d 1 3 }
: 39 y A K 7 4]
Gw, ہو‎ $e PAS Ka?
« o y A EA سی‎
H WM:
d - el
Lis ci Tic Edd Ea NS
pu DELE (1 SR o? E E de E elle
را ار روا‎ E e oU E E EF ip iod ret Ê $

TUNE adis di f
iat tilt 10s EE OE Aa Ok qo pud AA St KS: Sch 0 کی‎
d en aie Ga? ; Ws qii ARA sh bg: کی‎ Il»
icem con oca یہ‎ FATTE SE aM a Aën Lab sët? TE y 2
ere A hie Ree EE UR ا‎ E cR pd ا ا‎ it, a
DEA a te OM e LED e up kene st E D PERA DUC dau
^d $e tele? 1 CAE 0 ES AR HUS 1 Se E Zë i Abt 9 ا‎ eg, Ser F T^
HE PERADO نی‎ ch RC eti tcp oda ES EOS Pat ed a DEOS i
- betr وی‎ eh A E: uat eeh Pos ا ا‎ Jp US EA

Ee EE

EW by Lic A

D
te

z

De

INDEX

to terms defined in the Preface (numbers refer to paragraphs).

6, 15, 18

10

34

12

37

Focus

Free surface

Front of wave ~

Gravitational waves

Great undulations ,

Height of wave .
Homoseist . e

Horizontal earthquake

Inertia , ` `
Intensity ۰ ٠
Isoseist . e F

Length of wave e

Limit of elasticity .

Macroseismic . 4
Meizoseismic line .
Microseismic . .
Modulus of elasticity
Molar movement .

Molecular movement

Molecule , = ^
Normal wave . |. +.
Onduloso ^ E
Orizzontale e e

© 575 72
e 88
geg,
6, 13, 18
e 53, 72
. 28
e 51
. 33
` H
ek SE SW
. +
" 12
PAPE
CUM ME
e 64
e 64
۰ 18
(Kar
. 80
e 87
» 37, 80
. 9
e 9
. 28
٠ 27
e ag ا‎

—, sympathetic

Acceleration

Aftershocks

Amplitude

m measurement of

*

secondary

Angle of emergence.
Apparent velocity .

Area of propagation

Bulk modulus. à

Centre

Component . .
Condensational wave
Core

Cryptoseismic A

Destructiveness 3
Destructivity . -

Distortional wave .

Earthquake- motion

Earthquakes, cause of
——
—, types of

Elastic wave

Elasticity e .
Emergence . e
E picentre ۰ B

Epifocal area

XXIX INDEX.

Palpitating shocks . e ay Transverse waves . ٠ e `" aS
Particle, wave- . e " 2 True velocity . e 3 e 51

Path of wave . E e e 36

Period . ^ x : : 54 Undulation, see wave.
Pleistoseismic area . GE ig 32 Undulating shocks . ° 77ھ‎
Preliminary tremors . . e 39, 47
Propagation, area of ری‎ 33 Velocity of transmission . 25, 50, 69
, rate of : S 15950 — — of wave-particle . e 55, 72
Violence . . e . e 64
Range of motion, ° e 13

Volcanic earthquakes e e 2
Rate of propagation . 25, 50, و6‎

Vorticose shocks e e 7ئ2‎
Rayleigh’s waves, Lord . 5 19

Refraction e e e . 25 Wave amplitude ^ 7 6, 13 18

Resultant ~ š . e 35

, condensational ` ےا‎
Rigidity E e e . 9 ——-, distortional , 2 . 18
Rockfall earthquakes . 7. e 81 elastic. . > . 9

Rossi-Forel scale . e . 62 ,front . . . - 36

Secondary earthquakes . 2787 و‎ gravitational e ۰ 4
Seismic area œ e . 29254 ——, height e . - 6
Seismograph +. à 0 ——,length.. . e 6,15, 18
Seismometer . - : e, 06 =, particle, . e . 9
Seismoscope . ہم‎ 6 y paths Pia EL مر‎
Semiphase > ` 2.757, 345358 سے‎ propagation Es e 25; 50
Surface waves. ` . 4 10,49 , Lord Rayleigh’s . e 19
Sussultorio : . . . 37 —— , refraction ` . e 25

Sympathetic shocks . S . 92, 95 we, surface e e 45 I9, 49

æ

THE GEOLOGICAL SURVEY OF INDIA, -

MEMOIRS

OF

THE GEOLOGICAL SURVEY OF INDIA.

Report on the Great Earthquake of 12th June 1897, dy R. D
Oldham, A. R. S. M., F. G. S., Superintendent, Geological
Survey of India.

A ek

CHAPTER I.—INTRODUCTORY.

At about quarter past five in the afternoon of the 12th June 1897,
there burst on the western portion of Assam an earthquake which,
for violence and extent, has not been surpassed by any of
which we have historic record. Lasting about two and a half min-
utes, it had not ceased at Shillong before an area of 150,000 square
miles had been laid in ruins, all means of communication interrupted,
the hills rent and cast down in landslips, and the plains fissured and
riddled with vents, from which sand and water poured out in most
astounding quantities; and ten minutes had not elapsed from the
time when Shillong was laid in ruins before about one and three-
quarter millions of square miles had felt a shock which was every-
where recognised as one quite out of the common.

The earthquake found the geological survey in a manner un-
prepared for it. Of the officers available for despatch into the
earthquake shaken tracks, there was but one who had paid any
special attention to the subject of earthquakes, or had any knowledge
of the nature of the observations required, beyond such as might be
obtained from the ordinary curriculum of a geological student. The

B | ea

2 OLDHAM: GREAT EARTHQUAKE OF 1897.

area affected was, however, too large for any one man to examine
single handed, and the advent of the rainy season rendered it
necessary that the work of examination should be undertaken at
once. The four officers at headquarters, Messrs. T. D. LaTouche,
H. H. Hayden, E, Vredenburg and G, E. Grimes, were accordingly
despatched, with instructions to collect what information they could.
Mr. LaTouche was entrusted with the examination of the Assam
valley, and was subsequently instructed to visit and make observations
at Shillong, Cherrapunji and Sylhet; Mr. Hayden examined the line
of railway running north from Calcutta to Darjiling, and several of
the more important towns in Northern Bengal; Mr. Vredenburg
undertook the examination of the country to the west of this; and
Mr. Grimes made a tour through Eastern Bengal and the Cachar
valley. These investigations were confined to those districts where a
considerable amount of damage was known to have been done,
Mr. P. N. Bose was afterwards despatched to the eastern deltaic
districts, from which the accounts, received at first, were meagre, and
indicated a much less degree of violence than was at the time
intelligible. The reports submitted by these officers, so far as they
have not been incorporated in the text, will be found in Appendix A.
Every attempt was also made to obtain information by means of
letters and circulars; a large number of volunteer observers were
interested in maintaining a record of the aftershocks ; while the
Local Governments and the Imperial Telegraph Department had
been instructed by the Government of India to render all informa-
tion and assistance in their power.

It is on the information obtained by these means, supplemented
by a tour through the epicentral tract, which was made by the writer
during the cold weather of 1897-98, that the following report is
based, In accordance with the orders of the Government of India,
it deals with the earthquake from a scientific point of view alone,
and, except incidentally, the human aspect of the occurrence is not
dealt with. No tales of hairbreadth escapes, of deaths, or rescues
from death are recorded, nor is any reference to the manner in

2D

INTRODUCTORY. 3

which the disaster was met and the complete interruption of com-
munications remedied. These matters are dealt with in the official
reports of the Governments of Assam and Bengal, in the Adminis-
tration Reports of those Governments, and in the annual reports
of the Imperial and Provincial Departments which were affected.
Besides these an excellent summary of this aspect of the earthquake
has been given, in an easily accessible form,! by Mr. H. Luttman-
Johnson, formerly Commissioner of the Assam Valley Districts.

In the notices of the earthquake which have been published
previous to this, it has been described as the Calcutta, the Assam,
the Indian, or merely the, earthquake of 12th June 1897. The first of
these titles is obviously inappropriate, for Calcutta was nowhere near
the focus, nor was the earthquake so severe there as at many other
places. The second title, though better, seems still too limited, for
the province of Assam covers but a small proportion of the area
affected by the earthquake, nor did the focus lie entirely beneath the
area covered by the province of Assam. The term Indian is better,
for, not only was it felt over a large part of India, but the greater
part of the whole area over which it was felt, falls within the limits
of the Indian Empire. Even this name is not, however, altogether
unexceptionable, for more than a quarter of the area over which
it was felt lies outside the limits of the Indian Empire, while not
more than about a third can be said to lie within the limits of India
proper.

In fact, it is not possible to adopt any geographical name which
is altogether satisfactory, nor does this seem necessary in the case
of the greatest earthquake of which we have any historic record.
On the title page of this report I have referred to it simply as the
Great Earthquake of 12th June 1897, and this appears to be sufficient
to distinguish it from other, and lesser, earthquakes.

In dealing with the earthquake, I shall first give a selection from

the numerous accounts received by me, which will indicate the

1 The Earthquake in Assam, Journal of the Society of Arts, XLVI, 473-493 (1898).

B2 Dra)

TT SG lee o ہے میں‎
Wd me ^

4 OLDHAM: GREAT EARTHQUAKE OF 1897.

nature of the shock at different places, and to some degree its extent,
and shall then devote a series of chapters to considering the special
aspects of the data that have been collected. The course of the
investigation is sufficiently indicated by the headingsof the successive
chapters, but before proceeding to its description I must take this
opportunity of expressing my obligation to the numerous correspond-
ents whose names are mentioned in the succeeding pages,and to
those many others whom it has not been possible to name indivi-
dually. The information that each could give naturally varied much
in value and importance, but, such as it was, it was willingly given,
and what is more, subsequent correspondence and questions were
freely answered in some cases, I fear, not without considerable
trouble. But for this assistance readily rendered the information

regarding the earthquake would necessarily have been very
incomplete.

A

CHAPTER IIL.—NARRATIVE ACCOUNTS OF THE NATURE
OF TEE SHOCKS

In accordance with the general scheme of description proposed
in the last chapter, we cammence with the descriptions of the earth-
quake at Shillong, and of these I will first reproduce that of Mr.
F. H. Smith, of the Geological Survey of India, who writes:

<“ Some tremors are said to have been noticed for a few days previously by
sensitive persons; but if actually perceived they must have been very slight
indeed. On June 12th the severe shock commenced without any warning what-
soever (some people say they noticed a rumbling sound for to or 15 seconds before
the shock).
“I was out for a walk at the time, and was standing on the road which passes
the foot of the filtering tank of the Shillong
Personal experience, waterworks, near the school. At 5-15 (aceording to the
ordinary Shillong time, but it is not known how this
compares with absolute Calcutta or Madras time) a deep rumbling sound, like
near thunder, commenced, apparently coming fromthe south or south-west,
followed immediately by the shock. The rumbling preceded the shock by about
two seconds where I stood, and the shock reached its maximum violence almost
at once, in the course of the first two or three seconds. The ground began to
rock violently, and in a few seconds it was impossible to stand upright, and I had

(749

NARRATIVE ACCOUNTS. 5

Lo sit down suddenly on the road. The shock was of considerable duration,
and maintained roughly the same amount of violence from the beginning to
the end» It produced a very distinct sensation of sea-sickness. ‘The earth move-
ment was exceedingly sudden and violent. The feeling was asif the ground
was being violently jerked backwards and forwards very rapidly, every third or
fourth jerk being of greater scope than the intermediate ones.

“ The surface of the ground vibrated visibly in every direction, as if it was
made of soft jelly ; and long cracks appeared at once along the road. Thesloping
earth-bank round the water tank, which was some 10 feet high, began to shake
down, and at one point cracked and opened out bodily. The road is bounded here
and there by low banks of earth, about 2 feet high, and these were all shaken
down quite flat. The school building, which was in sight, began to shake at the
first shock, and large slabs of plaster fell from the walls at once. A few moments
afterwards the whole building was lying flat, the walls collapsed and the corru-
gated iron roof lying bent and broken on the ground.

* A pink cloud of plaster and dust was seen hanging overevery house in
Shillong at the end of the shock.

“My impression at the end of the shock was that its duration was certainly unde"
I minute, and that it had travelled from south to north. Several other observers
agree with me in limiting the first and great shock to 40 or 50 seconds. I have
heard since that its duration was timed roughly in Shillong and Gauhati, and put
down at 23 minutes ; but I feel convinced that this includes subsequent tremors,
which lasted some time. The violence of the shock may be imagined when it is
stated that the whole of the damage done was completed in the first 10 or 15
seconds of the shock,

* The buildings of Shillong might, before the earthquake, have been grouped

into 3 classes, which correspond now to 3 degrees of ruin :

Damage done, “1. Stone Buildings.—It is not too much to say that

every bit of solid stone work in the neighbourhood ot

Shillong, including most of the bridges, is absolutely levelled to the ground. The

stone houses, and conspicuously the church, are now reduced to flat heaps of

single loose stones, covered with torn and burst sheets of corrugated iron — the
remains of the roofs.

“ The walls do not show the slightest partiality in their direction of falling.
The stones have in every case been shaken loose, and have collapsed equally on
both sides of the line of the wall.

* Heaps of stones along the roads, broken for mending purposes, which stood
I foot high before the shock, are now flat, roughly circular patches 3 or 4 inches
in thickness.

* Two tall monuments of excellent cut stone work, about 20 or 30 feet in
height, are in ruins ; though in each case some feet of the masonry at the base
still retain an upright position—the individual stones being shaken from each other.
The ruins are scattered most impartially on all sides in a rough circle. The
pinnacle of the larger (Quinton) monument has been thrown down bodily, and
' lies some feet from the centre of the stone work pointing N.-N.-E.

“2, Ekra-built Buildings.— A wooden frame work, with walls of saz grass

LUN)

6 OLDHAM: GREAT EARTHQUAKE OF 1897.

covered with plaster. About half the buildings of this description are ruined in the
same way as the stone buildings. All thelarge ekra buildings are utterly ruined
inside, the chimneys in all cases being of stone work, the whole of which has fallen
with the plaster from the walls, and in many cases the roofs also.

* Small outhouses and villages of ekra-work have in some cases escaped with
the loss of the plaster. Some of the new larger buildings would also have escaped,
but for the stone chimneys, which have in every case wrecked the house.

"3. Plank Buildings.—Built on the “log hut" principle, a wooden frame
work covered with planks, resting unattached on the ground. The only buildings
of this description were stables or outhouses. In every case they have escaped
untouched, except where the supporting stone work has been shaken away, when
they have been slightly displaced.

** Trees in Shillong have not suffered much, two only were brought down by

the shock, and as they were leaning over to begin with, their direction of falling
does not give any evidence.

** [n Shillong itself the roads and hillsides are cracked in all directions, but the
cracks merely show the lines of weakness in the ground,
which has been simply shaken down by the violent
vibration, which has apparently acted in every direction.

* Small banks of earth have been flattened everywhere, and the band of
the artificial lake—a bank some 150 yards long and 30 or 4o feet high, made
mostly of earth, gave way almost at once when the great shock began. The centre

half of it has been carried bodily away down the valley below by the rush of
water. |

Country round,

* On the hills round Shillong four or five considerable landslips can be seen.
Patches of hillside debris have fallen from steep mnala banks, carrying trees and
undergrowth with them, and marked now as patches of red soil on the hiil—the
largest is about 300 feet in width and the same in height."

Another account which may bereproduced is that of the Assis-
tant Superintendent of Telegraphs, Shillong Division, contained in his
official letter No. 1473, dated 12th July 1897, Shillong, and addressed
to the Director General of Telegraphs. According to this, —

* The shock which levelled Shillong was immediately preceded by a milder
shock which did no damage, this was of a greater severity than the shocks usually
experienced in Shillong. This milder shock seemed to die away, but without
absolutely ceasing, it suddenly developed into a most destructive earthquake, which
no building or structure withstood for more than a couple of seconds. The
rumbling which accompanied the first shock increased in vclume as the shock
increased in severity. By that time I was outside, between my own bungalow
and another one not 3o yards from it. The noise ofthe earthquake was so great
that I could not hear the falling of the stone chimneys in either of the two
bungalows, nor the noise of the fall of part of my bungalow. As soon as the

( 6.)

NARRATIVE ACCOUNTS. 2
shaking was over, I looked at my watch, and allowing for the error of my watch
by that day's time, found the time was 5-16 P.M.

I estimate the duration of the whole earthquake to be about 13 minutes.

Duration and direction The first shock lasted about Io to 15 seconds, so that
of shock. the destructive shock lasted from 75 to 80 seconds.

The direction is more difficult to determine. Judging from the rumbling
sounds, they appeared to come from the south. During the shock there were
no undulating waves; there appeared to me to be no vertical motion, The hill
I was on atthe time simply felt as if it was being rapidly moved in a horizontal
plane backwards and forward. This motion was so violent that I was unable to
stand, but had to crawl on my hands and knees and hold on to a tree for
support. ,

Iexamined several stone structures, the Church, Telegraph Office,
Divisional Superintendent’s and Sub-Divisional officer’s offices, and found
they had simply collapsed, owing to the stones being shaken out of their position,
the debris remaining all round the site, with a slightly greater accumulation on
the north side. The roof in all these caseshas simply fallen almost exactly over
the place where the supporting walls were. The Quinton Memorial, which was
a stone spire, has collapsed, and the stones it was built of are simply lying piled
round the base. The stoneware column which was at thetop was lying south.

south-west and north-north-east. This I take to be the true direction of the
destructive shock. "7

One more extract, from an account given by the correspondent
of the Proneer newspaper in its issue of 3rd July 1897:

“ As regards the movements of the earth’s surface, I can furnish only the
personal impressions of one who has had little, if any, scientific training. The
movements seemed tome to be distinctly horizontal and undulating, the surface
of the earth presenting the aspect of a storm-tossed sea, with this difference that
the undulations were infinitely more rapidthan anyseen at sea. The depth of
each wave seemed to me to vary from one to three feet, but this is the estimate
of an unscientific individual who had never observed any similar phenomenon,
and your readers must take it for what it is worth. I should also add that the
earth’s surface seemed to me to have a gyratory motion, as though each little area
were moving in a whirl. It may be that the horizontal and gyratory movements
cannot take place at the same time, and that the latter was apparent only, the im-
pression formed in my mind being due to the rapid undulatory movements.” |

I have also received a number of accounts, both written and
verbal, all of which agree in asserting that there was a very marked
undulation of the surface of the ground. Different accounts place
the length of these undulations at from 8 feet to 10 yards and their
height at from 1 to 3 feet. It is probable that the highest of the

; (7 j

8 OLDHAM: GREAT EARTHQUAKE OF 1897.

former and the lowest of the latter values are nearest the truth, and
we may take it that the waves were about 30 feet long and 1 foot
high on the average, but that many may have been both shorter and
higher. All the accounts also agree in showing that there was a very
considerable vertical component in the shock, loose stones lying on
the surface of the roads being tossed in the air “like peas on a
drum,” this vertical movement being accompanied by a more marked
backward and forward movement of the ground, the sensa-
tion produced being generally expressed as being “ shaken like a rat
by a terrier? As will be seen in Chapter V, the range of this
motion must have been at least 8 or 9 inches. As to the period, or
rate of repetition, 1 was unable to obtain any really satisfactory
evidence; in the face of so great and so unexpected a catastrophe, it
could not be expected that attention would be paid to a matter which
would not be regarded as important, but, by asking people to try and
recollect their sensations at the time and to indicate the rate and
extent of the movement, I found that there was a very substantial
agreement as to the first, and that the rate must have been approxi-
mately one second for the double movement forwards and back-
wards. It may have been a little more rapid, but I do not think it
would possibly have exceeded 13 second. |

All the accounts from the station of Shillong agree in stat-
ing that the first indication that was noticed was a rumbling
sound, like that of a rapidly moving cart; but Mr. R. S. Strachey,
who was riding at the time on the Peak, some 1,400 feet above
Shillong, states that his attention was first attracted by the rustling
of the leaves on the trees, and that the tremulous movement which |
caused this lasted for several seconds before he heard any other
sound.

Passing westwards, we have the very interesting report by Captain
A. A. Howell, Deputy Commissioner of the Garo Hills, of the earth-
quake as felt at Tura, of its effects, and also of the manner in which
the administration of the district was carried on after the shock.

29)

NARRATIVE ACCOUNTS. o

PART I—TURA!.—

t t, To make this part of the report clearer, I have begun with a short dese
cription of Tura station, of which, too, a map (not printed) is herewith sent. In
this map, only the portion south of the Ronkon stream has been accurately
drawn. The contours are at 5 feet intervals. The rest was surveyed with a
chain and compass, the contours being roughly put in, merely to show the
general formation of the ground.

* 3. On referring to this map, it will be seen that the station is a very small
one, and built mainly on three spurs running from east
to west. The station has an elevation of about 1,300 feet
above sea level. It is situated at the foot of the western
end of the main range of the hills in this district, which rises immediately and
steeply above it to a height of some 4,000 feet above the sea, and thence continues
almost due east to the ranges of the Khasi Hills. The conformation of the spurs
consistsof a backbone of sandstone covered more or less deeply with a loose,
reddish sandy soil, plentifully mixed with boulders of (what I am told is) gneiss. ?
The crests of the spurs are broad and fairly level, but the sides are steep and
occasionally precipitous. At the bottom of each ravine isa small (hill) stream
flowing over a rocky bed.

d: The week preceding the earthquake began with fine weather, a clear atmos-

phere, and not much rain. On the roth there was a
ree LE Gë ze change. Heavy masses of cloud gathered round the
top of Tura pahar Ge, the high hill above the station),
and the air had a feeling of thunder in it, though there was norain until noon
of 12th, when a smart shower fell. This cleared off about 2 P.M., but was follows
ed, after a vivid sunset, by heavy downpours lasting until the morning of the
I3th.
“ 4. There being no telegraph wire to Tura, there is no means of checking
۱ Our local time. We depend chiefly on a sun dial, but as
The time of the earth. . ; : :

e it was inaccurately put up, the result is unreliable. The
guard-room clock stopped at 5-5 P.M., and this was the

time of the first shock by my watch.

“5. The exact time the earthquake lasted was not noted by anybody in Tura.
The consensus of opinion fixes it at five minutes or
longer. The shock began with a sharp vibration, in-
creasing in about 20 seconds toa violent rocking and
heaving motion of the earth, which lasted about three minutes; in about two
minutes more it gradually subsided to a gentle but distinct tremor. A few
persons assert that the shock was preceded by a loud rumbling noise, but this is

Short description of
Tura.

The duration of the
earthquake,

1 Reprinted, with a few trifling omissions, from the official report of the Assam Govern-
ment, published in the Assam Gazette and also separately.
2 The hills are composed of gneiss and other crystalline rocks, on which in places are
patches of tertiary sandstone and shales,— R., D, O,
(9 )

IO OLDHAM: GREAT EARTHQUAKE OF 1897.

doubtful (personally I did not hear it). Everybody, however, agrees that such
a sound was audible during the shock.

“ This tremor of the earth continued almost without intermission through the
night of the 12th June, increasing at irregular intervals
of some minutes to a strong and jerky shake of sufficient
force to make houses rattle and trees quiver. These shocks wete, and still are,
almost invariably announced by a distinct rolling sound as of far-off thunder.

* Nobody has kept 2ny record of their number, but from several hundreds
in the twenty-four hours, they have decreased to ten or twelve a day; even now
some are nearly as strong as those immediately following the first and severest

Subsequent shocks.

shock.
« 6. The effects of the earthquake, as seen on the ground, are (1) landslips

Effects of the earth- (2) cracks and fissures; (3) ridges and furrows; (4)
quake on the ground. depressions of the surface.

(1) Landslips.—Landslips ate now visible from the station in every direction,
but are fewest to the north. They have occurred wherever the ground was steep
and not formed of solid rock. In the station itself the southern spur has suffered
most, and more on the south side than on the north. There isa large slip at the
head of the nulla to the south of the Deputy Commissioner’s bungalow. From
this point, as far as the parade ground (on the western extremity of the spur, not
shown in the map), the whole hillside has slipped more or less; on the northern
face the landslips are both fewer and smaller. The small spur on which the
inspection bungalow and charitable dispensary stand displays similar effects. On
the south side, half the knoll on which the clerks’ houses were built has come
down, and there is a smaller slip within a few feet of the charitable dispensary.
On the northern face, only a few rocks have fallen.

« The large spur on which the cemetery and Forest Officer's bungalow are
situated has small landslips on both sides. The spur on which the mission come
pound lies has suffered least of all, there being only one small slip on its southern.
face. On the other hand, the high hill immediately behind it is scarred all over.

(2) Cracks and fi ssures.— Cracks in the ground are to be seen everywhere, but
chiefly in the vicinity of landslips and along the upper edges of sloping ground,
Their direction follows the line of the hills. Besides them, however, there area
good many on the level crests of the spurs far from any sloping ground and

running in various directions. In some, a walking stick can be pushed down to

a depth of 2 or 3 feet. "
« (3) Ridges and furrows.—In the western portion of the southern spur and

all round the Civil Surgeon’s quarters to the distance of a mile down the Man-
kachar road, where the soil is
sandy and the surface fairly
level, the ground looks as if a
steam plough had passed over
it, tearing up the turf and throw-
ing the clods in every direction,
Fig 1. some up hill and some down, and

in many cases turning the sods completely over, so that only the roots of the grass

(09

NARRATIVE ACCOUNTS. 11

are visible. In other places, the earth seems to have first opened and then closed
again with great pressure, forcing up the turf along the line of cleavage into
ridges, thus (Fig. 1).

In other places, again, large pieces of turf are found lying on the grass with
no mark or sign to show the place from which they had been torn,

*(4) Depressions of the surface.— Depressions of the surface of the earth are
only a few inches deep and apparently of small area. Their edges at first were
clearly defined, but now have become worn down by the weather and by the feet
of men and animals. Those near sloping ground are due, no doubt, to the
slipping of the earth or displacement of some large boulder below, but others
occur in the centre of the most level places, and indicate perhaps an actual
subsidence of some substratum of the spur itself.

** Trees have not, as a rule, been damaged, except on steep hill sides and in sandy
soil, where in many cases the earth seems to have been shaken away from the roots.

* A stone onthe parade ground, about 20 inches in diameter and 10 inches
thick, partly imbedded in the ground, was thrown to a distance of about 3 feet,
and turned upside down. A cylindrical metal rain-gauge on the Deputy Com.
missioner’s lawn was also similarly displaced.

* 7. All buildings in Tura, both Government and private, may be divided into
four classes :—First.—Those built on wooden piles from
10 to 20 inches in diameter, driven several feet into the
ground and reaching up ٥ the roof. These have a raised board floor, mud and
reed walls supported by timber battens, and a roof of thatch or corrugated iron.
Of this class are the houses of the Deputy Commissioner, District’ Superinten-
dent of Police, Forest Officer, and Sub-Engineer, the inspection bungalow, cut-
cherry, and the houses of the American Baptist Mission.

* The houses of the Deputy Commissioner, District Superintendent of Police,
Forest Officer, and Sub-Engineer are damaged more or less severely inthe same
fashion. Several posts have sunk from a few inches to a foot deeper into the
earth, causing the floor to buckle and the roof to sag. Many, too, are out of the
« perpendicular. At the point
each post enters the ground, a
cup-shaped depression, from
one to six inches in depth and
diameter, has been worn round
it thus (Fig. 2) as though the
post had been given a circular
movement :

«< Many of the timber bat-
tens supporting the walls have
been wrenched loose, most of
the plaster has fallen, and, where old and rotten, the reeds have come down
with it. The bungalows of the Deputy Commissioner and District Super-

[1

Damage to houses, etc.

Fig. 2.

t2 OLDHAM: GREAT EARTHQUAKE OF 1897.

intendent of Policehave suffered most. The former was built in 1872«73, the
latter in 1875. The Forest Officer's bungalow was built about twelve years ago.
Its walls have been badly damaged, and to a smaller extent, also the posts, floor,
and roof.

* The Sub-Engineer’s house was built in 1883-84; it has suffered in the same
degree as the Forest Officer's.

* The inspection bungalows and cutcherry were erected in 1889 and 1889-9o,
respectively. They are both in fair condition as regards posts, floor and roof,
but the walls have been alniost destroyed.

* While Government buildings havé fared so badly, itis curious to note that
the houses ofthe American Baptist Mission have withstood the shocks with com-
parative immunity. The reason perhaps is that the posts, instead of being driven
deep into tlie ground, mostly rest on stones almost flush with the earth, or flat on
the earth itse!f. Their walls, too, on the sides exposed to the weather are protected
with stout planks, which no doubt increase the strength of the structure.

“ The second class of buildings differ only from the first in that they have a
stone and cement plinth, instead of a raised board floor. In this class are the
Civil Surgeon's house, the charitable dispensary, the post office, the treasury,
and some shops in the bazaar. The most remarkable point about them is the
manner in which the plinths have cracked and broken up. It is most noticeable
in thetreasury. This building stood on a gentle slope a few feet from the top of
a steep hill-side. It had a strong frame of timber, walls and roof of corrugated
iron, and a massive plinth of cement and large stones. The plinth was 7o feet
long, and 32 feet 6 inches wide with a maximum and minimum height of 8 feet and
1 foot 6 inches, respectively. The floor was of concrete and cement to a depth of 9
inches. Solid as it was, it crumbled away like a pile of sand before the first
shock had stopped. The stones rolled cut on all sides, and the floor was broken
into little pieces. The whole building tilted over towards the khud, wrenching
the main posts in every direction, and uprooting or breaking the smaller ones
which supported the verandah. The plinths of the other buildings aiso cracked
up, and the sides fell away, leaving gaps under the walls, but they do not show
Signs of such violent action as the treasury. The frame and roofs are almost all
sound; in the Civil Surgeon's house only, the thatch has parted along the ridges
of the roof and slipped tothe ground. The walls have been denuded of plaster
and the battens have in some places been loosened. These buildings were erected
as noted below:

Civil Surgeon’s quarters built in 1897. Post office built in 1887-88.
Charitable dispensary 49 99 1800-01. “Treasury 55». 5 16097389;

<“ The third class consists of buildings with the ground for the floor, timber
or bamboo frames, walls of mat or reeds lightly plastered over, and roofs of
thatch or corrugated iron. The out-offices of the different bungalows belong to
this class. Being old and badly built for the most part, they have been reduced
to ruins, Other buildings of this class, consisting of the police barracks, clerks?
quarters, and the majority of the houses in the bazaar have stood well. Some

ER,

NARRATIVE ACCOUNTS. 15

have collapsed, owing either to their original ricketty condition, or to the ground
having fallen away under them.

“The fourth class consists of the Garo's houses with a raised floor and built

entirely of wood, bamboo, cane, and thatch.
. “These houses, when on level ground and in fairly good repair, have not been
at all affected by the shaking. Unfortunately, many of them were built (in a hilly
district like this) on steep hill-sides, or at the base of cliffs, and numbers of them
have been destroyed by landslips, In all the houses in the station, furniture was
thrown about in great confusion, pictures fell, the pendulums of clocks were
shaken off, chairs and tables shifted about like grains of ricein a winnowing
basket. Almirahs and book cases were overturned; in short, no moveable
article, however big and heavy, was left in its original place.

* 8. The station roads and bridges had been badly damaged; that part of the
road running past the District Superintendent of
Police’s house, marked A B C in the map, has slipped
away altogether, and the path from the jail tothe point marked D is covered
by a landslip, The stone abutments of the bridges over the Ronkon and the
Mafilkol have been shaken down, but the bridge over the Rengri has escaped
untouched.

“Tosum up. The damage done by the earthquake has been greater on the
southern than on tke northern sides of the spurs, while it is also most noticeable
on the southern spurs, decreasing gradually towards the north. The small collec-
tion of houses near the mission compound, marked Chandmari in the map, shows
no signs of any disturbance, nor do the bridle-paths to Rongrenggiri aad
Salmara for the first 3 miles from Tura. On the other hand, the cart-roads to
Dalu and Mankachar are blocked with landslips right up to the station,

Station roads and bridges,

PART II.—THE DISTRICT.

“ o, Owing to the difficulties and the deficiencies in the means of coniu cica-
tion in this district, full reports of the damage done have not yet been received
(rom all sides.

“In the plains portions of the district deep cracks and crater-like pits appeared

in the ground ; one of tbe cracks is as much as a mile

Damage in the plains, long, 2 or 3 feet wide, and 16 feet deep. The pits

average abovt 6 feet in diameter, though one is said to

be as much as 4o feet across. Through these fissures, jets of sand and

water were thrown up to a height of 6 or 7 feet, in a few cases accompanied by

pieces of coal, peat, resin, masses of semipetrified timber, and a black earth

hitherto unknown in those localities. The bed of the Ganal was upheaved 20 feet,

and so was the bed of the Jinjiram, but not so much. Both have since settled
down again, though not to their normal level.

* Many houscs sank into the grcund bodily, the roof alone being visible, but
no loss of life has yet been reported. Several villages were, and still are, partly
submerged, and some thousands of bighas of cultivated land have been made

Ee]

14 OLDHAM: GREAT EARTHQUAKE OF 1897.

useless. Of the crops, jute has suffered most; in some fields plants as high as a
man have been completely buried in sand. The ashu crop has suffered a great
deal less, and, where not destroyed, promises to be up to the average. Several
bhils have silted up, and the newly-formed land is now being ploughed for sali;
as the soil is excellent, good results are expected. The earthquake caused a
regular panic amongst the plains people. Whole villages were deserted for days,
while the inhabitants took refuge inthe hills. A few Garos took advantage of
this to loot their granaries. Considering, however, the extraordinary facilities for
thieving afforded, very little crime was committed.

* şo. In the hills, landslips have caused a great deal of damage, granaries
have been wrecked, and much grain lost. The cultivated
slopes of the hills have cracked so as to injure the crop,
and in some places the earth has been so severely shaken as to destroy it.

* It is generally stated that some hills have sunk bodily many feet while others
have risen. In the case of two hills between Tura amd Rowmari, this as-
sertion is corroborated by the signallers of the Military Police, who have been
constantly practising with the heliograph from these two places.

<“ 11, I regret to say there have been more lives lost than was at first supposed.

Up to date the number is 27, but as the reports arrive

Mortality. from the more inaccessible parts of the district, this

number will, 1 fear, be considerably exceeded. The

greatest loss has occurred in the lower part of the Someswari Valley in the circles

of Bong and Damjing Lashkars ; the latter with his whole family was buried in
his house by the fall of the hill above it.

“12. The Garos generally were thrown into a state of stolid bewilderment by

the earthquake. They left their fields, and retired

Attitude of the people. into their village houses to await further catastrophes.
The Garo belief is that the world is a square flat boay

hung up by a string at each corner. . There is a squirrel always trying to gnaw
these strings, but to prevent it a demon was appointed. This demon, however,
neglected his duty, and in order that his attention might not inthe future be di-
verted from his work, he was struck blind. Now that he can't see, the squirrel
of course has the best of it, and it is feared that when. one or two of the strings are
gnawed, the earth will be turned upside down." Another story is that Her Most
Gracious Majesty, not content with the last earthquake, has ordered another and
more vigorous one to be followed by a cyclone. ‘That it is in the power of the

Damage in the hiils.

Maharanito do so is never doubted. One men asked for a parwana to forbid `

the hill behind his house from slipping down on to him. Had the houses of the
European officials in Tura not been wrecked, the Garos would have made up
their minds without doubt that the recent catastrophe was the work of the Sahibs,
and excited by the wild stories in common circu'ation they might have given
some trouble. i

1 1 was informed that a number of Garo children, who were attending the mission school,
convinced that this had come to pass, rushed underneath the platforms, raised above the
ground, which formed the floors of the mission houses, so that, when everything was reversed,
they might still have a solid floor to stand on.—R. D. O,

E 14.)

NARRATIVE ACCOUNTS. 15

13. The district roads have all been badly damaged ; the cart-roads to Dalu

and Mankachar were impassable for weeks, except

Roads. by foot passengers. In the hills, they were blocked

by landslips, in the plains embankments sank 2 or 3

feet for long distances, breaches were frequent, and bridges had nearly all sub-

sided more orless. At the 22nd mile of the Mankachar road, a timber bridge,
80 feet long, was heaved up to a height of 8 feet above its proper level.’ ,

To the north, in the alluvial plain at foot of the Garo hills, but
still, as it would seem, within the area of the seismic vertical, we
have the account of Surgeon-Major E. F. H. Dobson, Civil Surgeon
of Dhubri, contained in a demi-official letter addressed to Mr. H. J.
. Cotton, C.S.I., Chief Commissioner of Assam.?

“On the 12th, I sent you a demi-official, and about four hours later, that is
at 3 p. m., left the station by the ferry steamer, which landed me on the opposite
bank at 4-15, and from there I drove to Jemadarhat, where I met my camp.
There was a stormy sky ahead and a strongish breeze blowing whilst I was cross-
ing the river ; on reaching the opposite bank, had a few drops of rain, and, on
the whole, had a cool and pleasant drive till I reached Jemadarhat, which is 31
miles from the opposite bank of the river. The resthouse at Jemadarhat was
burnt down some months ago, and my servants rigged up a small godown about
16' X 8' for me to stay in during the night.

* At 5 P.M. came this terrible earthquake. I invariably sit out an earthquake,
as I have always thought the worst is over before one has time to get outside a
house, but though I was safe enough in the light shanty I was in, I had to rise, as
the shock seemed never ending and a roar of water flowing near and the hum of
voices compelled my geting up. The villagers were panic-stricken and fleeing
from their houses to the Trunk Road. The ground I at once saw was everywhere
cracked up, sunk, etc. Behind the old rest-house is the termination of a very long
bheel. lt was a wonderful sight, this bheel. The water rose steadily in it, and
in two minutes must have risen at least twelve feet and then overflowed, and with
this overflow, the whole country, as far as the eye could see on each side of the
Trunk Road, was one sheet of water. The fact being that from all the innumera-
ble cracks water had gone on spouting, and that, coincidentally with the overflow
of this bheel, the water level had risen to the same height all over the surrounding
country,

* [n addition to the cracks from which the water welled up carrying the sand
with it, innumerable jets of water, like fountains playing, spouted up to heights
varying from 18 ‘inches to quite ڈو‎ or 4 feet. Wherever this had occurred, the
land was afterwards seen to occupy a sandy circle with a depression in its centre.
These circles ranged from 2 to 6 and 8 feet in. diameter, and were to be seen all
over the country. In some places, several were quite close togetber; in others
they were at a distance of several yards apart. They had the appearance of

2, Assam Government report, Appendix V.

Woche

16 OLDHAM: GREAT EARTHQUAKE OF 1897.

small craters I have heard it said that the water from the cracks and these soe
called fountains was warm, almost hot, but this must be due to some one's imagin-
ation: I had to wade through a good deal of water whilst the water level was
rising, and had the temperature of the water been abnormal, it must have made
itself felt to me. It must not be overlooked that the country had been water-
logged days before the earthquake came, and that it must have been this water
which was driven up with the subsidence of the land, as I believe ‘the quake was
due to a subsidence, i

* [n front of Jemadarhat resthouse the Garo Hills are within 5 miles, and
behind it is the Brahmaputra at a distance of not more than 2 miles. The Trunk
Road remained above water, but in certain portions, where it had subsided from
t to 2 feet, the water flowed across it, and at these places where the panic-stricken
villagers tried to run towards the direction of Lakhipur, which is 12 miles off in the
Gauhati direction, the poor creatures got more alarmed than ever, for they found
themselves sprawling about in the water, unable to realise what had barred their
progress, and not knowing if they were going to get out of it again. At all these
gaps or places, where the earth had subsided, or where fissures and cracks had
formed, the rush of water from them all carried with it an immense quantity of
sand, and this was what barred the progress of these panic-stricken villagers.
This sand had a particularly quicksandish feel about it, but fortunately, as a rule,
was not deep on the portions of the road that had subsided near the resthouse.

“The main shock seemed to me to have lasted nearly six minutes but perhaps
in this I am wrong and perhaps I may have missed a cessation and taken two or
more shocks as one, for certainly I noticed no cessation. It seemed as though
death from drowning must have been the end of every one near me. Although
the shocks went on being repeated incessantly at intervals of a few minutes, the
rise in the water ceased after half an hour, and then the idea present to one's eye
was that the country on either side of the road was nothing more than innumerable
sandechurs! of different sizes, such as are met with in the Brahmaputra. This
meant, of course, that crops were considerably covered by this sand. Most of the
excavations made by the earth-cutters annually on each side of the Trunk Road
to enable them to throw fresh earth onto it have been filled in by this sand, and
the Public Works Department will not have quite such heavy ratesto pay for
earth-cutting, as the leads will be comparatively short when they begin work
again; but for this, which is really no advantage to speak of, the loss to the pro-
vince must be enormous.

* All the panic subsided by about 9 o'clock; the hut I was originally in was
now under water, and Í could not get back toit. Many of the villagers? houses
and those of the Kayas had sunk a good deal, whilst others filled with two er
mo-e feet of san? through the doorways. My horse broke away towards the end
of the first shocks, and made for home, the syce after her as hard as he could go,
but on reaching a good strong wooden bridge, she was stopped by the bridge
subsiding on the Dhubri side. Cattle seemed to find great trouble in extricating

t i.e, و5‎ d- banks.

( 7694

NARRATIVE ACCOUNTS. 17
themselves from the sand and water, which but a few minutes before was their
grazing ground. Villages were wholly vacated and refuge taken wherever
high land was available. Almost everything is said to have its ludicrous side,
and in the midst of confusion and the marked panic of the villagers, it was
amusing to find my cook busy at work preparing dinner within half an hour of the
occurrence. There he was, as though nothing had occurred, on the plinth ofa
burnt down hut only just above water and surrounded on all sides by it. I was
very grateful to him later on for not having parted with his wits.

* As the rush of water ceased and order was to some small extent renewed,
those who were able to do so scrambled to their huts, and brought out boxes and
bedding, etc., etc., and, with their belongings, encamped on the road for the night.
At about 1o it struck me I might have a roof over my head if I got near the
ferry ghát, as there was a tumble-down serai there. A had gone half a mile when
two men assured me it was impossible for me to |progress, owing to a gap of
several feet in the road through which water flowed, and across which they had
just swum. On hearing this, I returned, and with my bed on the road, attempted
to get at some rest. This was impossible, with the continuous and oft-repeated
seismic shocks at short intervals, which, whenever the least bit strong, were
followed by a wail of human voices. Cattle, goats, and dogs formed a goodly
addition to those resting on the road for the night. At 2 A.M., a sharp drizzle came
down, and with my umbrella over my face and a rug over my body, I weathered
it out till 3 a.m. when my rug had got wet through. Close by was the resthouse
well. 1 now transferred my head-quarters to it, by wading across 2 feet of water
for about 15 yards, and then spread a dryish rug on the masonry platform of
this well. The well and its platform had both been badly cracked at 5 P.M.
and the water had risen to within 18 inches of the surface. It had a thatched
' covering over it, and this sheltered me during two hours of heavy rain.

“At 5 a.m. I rose finally, having had a fair night's rest under most extraor-
dinary circumstances. Having had Chota Hasree at 6, I started to walk back to
the ghát, having learnt that the road was impassable towards Lakhipur. As the
road was heavy and my syce had developed strong fever, I left him the trap to
drive in slowly. The whole Trunk Road from Jemadarhat to ferry ghát, which
I had seen, had cracks running along it, which were in some instances over
100 feet long; in others the earth had subsided for a similar distance, and
then, again, transverse cracks, depressions, etc., existed all along the road
at distances of afew feet for miles; the same thing seemed to have occurred
off the road side. The villagers stated that they had lost the grain they had
stored in their houses. On arrival at the ghát, it is scarcely necessary to say it
was much changed, the river having risen ¢onsiderably since 4 P.M. However,
after getting almost knee-deep in many places in very disagreeable soft sand,
about which I was hopping for quite half an hour, I got safe on to the ferry, and
returned to Dhubri at g a.m. Shortly before reaching the ferry ghat, my syce had
sent word that my horse was drowned, and desired me to return. At the ghat
another messenger brought me word saying the animal was well stuck in soft sand.
So I sent back coolies, and much to my surprise, found her none the worse, and at

C | (5594)

18 OLDHAM; GREAT EARTHQUAKE OF 1897.

Dhubri by 5 P.M. Itseems she had got into the sand at about 7 A.M. and
remained in it till 1 p.M. There were several gapsinthe road between Jemadarhat
and the ferry ghát, but none over 2 feet wide, and the syce instead of exercising a
little care when having to go off the road, took the shortest cut possible, with the
result noted by me. The huge gap I was told of the night before turned out to be
a myth.

“On returning to Dhubri, everything was as I had pictured it at 5 P.M. the
previous night. The town is in ruins. “It has sunk everywhere, and for hundreds
of feet in a stretch it is cracked and fissured. The dispensary (main building)
stands, but has suffered much. It is built on iron pillars, and its brick walls are
only about 12 feet high and 2 feet thick; above, its walls are lath and plaster.
Hence the reason it has stood. The cutcherry, the forest house, Local Board
bungalow, circuit-house, the club, telegraph office and telegraph signallers”
quarters, and the Gauripur zamindars’ house stand, also Leslie’s and Mabher’s,
because they are built on piles. Of course the plaster and lath are all gone.
The treasury, Jewett's house, Jolly’s house, the dak bungalow, the Bijni Hall, two
large houses at the immigration depóts, the liquor shop in the bazaar, and the
Muhammadan musjid and my house still stand, but are dangerous to enter and
must all be dismantled. The post office, Hallifax’s, Bogribari zamindar’s house,
the high school, the church, the Sikh temple, the new jail, the newly built magazine,
the seven newly built houses in the dispensary ground, and all the outhouses of
every single brick-built house are down flat. So far, only one fatal case has been
reported, and this a native child about 5 years of age. The natives are all
panic-stricken, are living on the roads, in boats, and in steamer flats, and seem most
helpless; the better classes encamp out on the roads during the night, although
several of their houses are safe. Several of them have, however, suffered consider-
ably ; houses, especially those built on new reclaimed land near the Dhubri back-
water, have subsided a good deal towards the back.

** Just beyond the front of the old jail and extending for nearly a quarter of a
mile below it and in the direction of the old bazaar, the land has subsided quite 5
feet, but this subsidence is not so great as we get further away from the jail ; in
fact it is as little as 2 feet. Unfortunately for the station this subsidence near the
jail is all but communicating with the river, and runs parallel to it behind a row of
houses situated in the old bazaar on the river bank. As soon as the river rises, it is
certain to flow into this subsidence (which already has water in it), and then there
will be one steady flow from the river right on into the recently-excavated tank in
connection with the new drainage scheme of the town. Shortly after this
takes place we shall, I think, find the narrow strip of land between this
tank and the main bank of the river will be zon est, and that the river will be
flowing through what was once the tank and continuing its onward course carrying
all before it and skirting uncomfortably near the new bazaar.

“The station drains are kutcha, and almost every drain is choked up with
sand carried up from the cracks up to or very nearly on a level with the roads.
1 he station tanks have risen considerably and some overflowed. In several of

them islands of sand are visible, clearly showing that, though the tanks are full,

DB

NARRATIVE ACCOUNTS: 19

the actual depth of water is not very much. This I have verified in some instances,
and it is further proved by the condition of the public wells in the station. Almost
all the pits have been levelled up by the sand from the cracks, The new bazaar,
which was very low ground, has benefited considerably by being raised with this
overflow of sand. It is at present also more or less flooded. This sandy overflow
was restricted tolow lands, more especially as bunded roads showed no sand or
water on them, till after they had subsided to a certain depth. The two bunded
roads in the station, vizs., the Bidyapara and Gauripur roads, and the two or three
branch roads from them, as also the 33 miles of road on the opposite bank of the
river over which I had travelled on 13th, have all subsided a good deal in places,
and I think it not a bad comparison to make when I say they are not at present
very unlike switch-back railway roads on a small scale.

** The station is terribly cracked up and has subsided in places, but this is not
to be wondered at, more especially when we recollect that during the rains it is
almost an island, and a very narrow one with rivers on three sides of it, and these
tivers forming three large expanses of water tearing along, The front of this
station is, as you know, very high land, a sort of hill, and on rock. This portion
was not cracked up, but as soon as we reach the first road behind the front or
river side road, which is low land, we notice terrible cracks in the ground both in
numbers and extent, and this road is not more than 300 yards at the back of the
river side or front road. It is in this high portion of the station that natives of all
classes seek refuge, especially as the evenings begin to close in. lam told the
Brahmaputra rose several feet at the time of the earthquake. Next morning
when I crossed it, it seemed a very different river: a huge sand-chur in front of
the station had disappeared ; it seemed in full flood, drift wood was plentiful, and
I noticed two small boats (empty) being carried along. Bilasipara and Lakhipur
zamindars country residences are down in a heap. Gauripur, Rupsi, and Bogri-
bari are said to be badly damaged. The Joldoba or Gara bridge and the Krish-
nai bridge, our two chief iron bridges, are reported broken. A hill between
Goalpara and Agia is said to have slid forward towards the road. The sál wood
railing of the bridge near Jemadarhat is quite 8" x 8", and it was split right across
on one side. On 13th night we had heavy rain. It had been fairly cool till
yesterday, and up to the present it is hot again. [t has just rained for five minutes
at 2 PN, and is threatening to do so again. But what about these continuous
seismic shocks, which are really strong at intervals of every half or one hour?’ A
severe one again at 2-40 P.M., and so on all day and night. We seem to have had
such a severe beginning and so much of it since that we are quite accustomed to
shocks that at an y other time would have been referred to as very serious.

* There seems to be almost one continuous tremor of the earth going on. I
attach a tabulated statement regarding the wells in the station, which, I think,
will be of great interest, It will reveal how they have filled with sand, and also
that those which had not filled entirely with sand filled almost to overflowing with
water. In some, both sand and water had overflowed. I am afraid this silting
up of sand in wells, tanks, dheels or jheels, and I presume in rivers, will mean

C 2 (39 ]

20 OLDHAM; GREAT EARTHQUAKE OF 1897.

great misery for the villagers and their cattle when the dry season sets in, and
especially if we have a long drought."

Though not within the area of the seismic vertical, the northern `
part of the Maimansingh district lies very close to its limits, and
from here we have an account by the late Mr, ۶. Mercer, Superin-
tendent of Telegraphs, who writes as follows, in a report addressed
to the Director General of Telegraphs :—

“On the 1zth June, I was at Attrabari in the Mymensing district, with
Mr. Parrott, Sub-Assistant Superintendent. We were stopping in a timber and
mat building belonging to the zemindar. Shortly after 5 r.m., Calcutta time, the.
building began to sway and creak, at first thinking it only an ordinary slight
earthquake we did not move; but as the motion increased we went out into the
garden; we had gone about five paces from the house when the ground rocked
violently and we were both thrown down, as we got up I looked at my watch and
noted the time 5 hours 6-7 minutes P.M., Calcutta time. The shock appeared
to come from the south-west, and on looking in the direction we saw a series of
earth waves approaching over the surface of the ground exactly like rollers on the
sea coast, as these passed us we had some difhculty in standing, but none of these
waves reached the intensity of the first, which had overthrown us. It was raining
at the time and I had no compass but by marking a clump of trees which
appeared to be directly in the line of the waves with the place where we were
standing, and checking it next morning at sunrise; l estimated that the shocks
were from almost south-south-west to north-north-east. As the waves above
referred to subsided the ground began to crack at our feet, a long crack appearing
at right angles to the general direction of the waves ; not knowing what further
surprises might be in store for us we left the garden and went out on to the main
road, as we reached it we saw a series of “ geysers ” springing up in some waste
ground on the far side of the road, pouring out sand and water, the highest of
these reached, I should say, about 4 feet; on seeing these I ran back into the
house to get my camera, in the hopes of getting a photograph of one, although
as it was raining slightly I had not much expectation of getting a good result. I
had some difficulty in taking a photograph as just as I focussed the camera on
the largest, a new one sprang up at my feet making it necessary to shift rapidly,
moreover these geysers varied in intensity considerably, one would spring up two
or three feet and then suddenly die down and another rise on a different spot. The
water I noted was not hot.

* * & * * * *

* On my way to Isvarganj and Ramgopalpur I noticed literally hundreds of
miniature craters where these mud and water geysers had sprung up, in one span
of 100 yards I counted 52 between the two posts in a strip of ground about 20 feet
wide, the largest had a dimension at the bottom of about 4 feet and was above
1 foot high.

( 20 )

NARRATIVE ACCOUNTS. 21

* At Isvarganj office I found that a large crack had opened in the ground
through the office, this was about five feet wide, and as far as 1 could judge
about 15 feet deep ; portion of the office had fallen into it. At Ramgopalpur I
found the office, which is in the lower floor of the gate house of the zamindar’s
house, was merely a heap of bricks and plaster.

“ All along the road into Mymensing where numbers of cracks and crevices,
some five and six feet wide, rendering the road impassable for wheel traffic, some
of these had opened on the site of a telegraph post, and the whole socket and
subplate could be seen, the post leaning up against the side of the crevice having
slipped down as far as the subplate would permit it to go.”

From the same district I have also a copy of a letter addressed
by Babu Hiranmoy Mukerji, of Muktagachha, to the Chief Secretary
to the Government of Bengal, which I reproduce as a clear and
simple account of the earthquake, besides expressing the views of
an important section of our fellow subjects on the cause and cure

of such catastrophes of nature,

* [ have the honour to submit the following report on the effects of earthquake
in the District of Mymensingh, so far as ] have been able to ascertain, for the
information of Government ; and to suggest certain measures for palliation of the
mischief done by the earthquake, for such notice as the Government will be pleased
to take of them.

“2. The shocks were felt at 5*6 in the afternoon of the 12th June 1897, and
their duration was about five minutes. The shocks were preceded by a very grave
sound like that of a distant thunder, and were so violent in nature that the earth
swayed like a cradle with all the contents upon it. The waving of the house-tops
above and the rending of the earth below were something very awfulto look at.
And so long as the shocks continued, nobody could stand on his legs, but had to
stoop down and cling to the earth as fast as he could with both the hands and
feet, expecting nothing but death at every moment. After the first quake was over
there was a temporary rest. Again, fromthe evening, the shocks were felt at
quick intervals during the whole of the night and the whole of the next day. And
even though the force was much abated, yet it was sufficient to create a good deal
of uneasiness amongst men for the safety of their lives. From the second day
the shocks continued with gradually less and less force, lasting for a minute or
. less, rocking the housesand things, and either making the smaller cracks in the

buildings wider and wider, or causing those collapse in which wider cracks were
made by the first shock. The shocks continued up to 20th July 1897, though of
course their number and force lessened every day.

** 3. As the effects of the earthquake are generally the same everywhere in
the district, I think it is unnecessary to describe what took place in every village
and town. It would be sufficient, I think, to give a general summary of the
whole.

fet.)

22 OLDHAM: GREAT EARTHQUAKE OF 1897.

“4. Firstly, as to the surface of the land. The ground was rent into two by the
violence of the shock. _In many places, especially in those on the eastern side of
the river Brahmaputra, large gaps were made from 4 to ro feet wide, running to
a considerable length, and resembling a canal, from which oozed out sands and
water with or without a considerable force. Within a short time, the canals so
made were full of water and were not fordable in many places. In some places
the gaps were from 1 to 3 feet wide, and their depth was from 10 to 12 feet or more.
Water appeared in these gaps a foot or less below the surface. In one and the
same place there were several gaps or fissures running crosswise, causing the
ground heave up in one place and sink in another. These gaps sometimes
passedthrough houses and buildings! causing tbe former to sink down with
posts and fencing, and leaving the thatches only above the ground, and the
latter entirely collapse. In some places, however, the houses and the buildings
were only partially buried. The roads in some places sank down to the depth of
several feet and in others heaved up. They were for some time almost impassable,
and up to this day very many nearly remain the same. In some of the corn-fields,
these gaps took the form of little rivulets, which ran sometimes to a considerable
length, while in others, sands were deposited to such-a height that the jute and
the early rice crops, which stood over them, were buried together with the hopes
of the cultivators, which were fastened upon them. In one place, a large tree sank
so deep that only its branches remained above the surface of the ground.

“5. The shock of the earthquake was very severe in that part of the country
which lay on the eastern bank of the river Brahmaputra. It may be noticed
here that that part of the country has got an infamy marked upon it because of the
Pandavas not having set their foot upon it. There as well as in few places on
the northern bank of that river, some shallow rivers, named in the margin were

dried up and sands were deposited on their beds, making

Kharia. them appear like roads newly made. Many tanks
ج‎ similarly dried up and sands filled up the hollow. This.
Rangsha. will again cause the scarcity of drinking water in those
tame: places. These gaps, fissures and the deposit of sands
e 0 elc. in some of the corn-fields have been so great that it

would be impossible for long time to come to bring these
fields under cultivation again, or perhaps never at all.. [n some places an entire
village has sunk, and water come upon the surface to the depth of the knee or
theloins. The people of such villages had to pass one or two days on crafts
made of plantain barks. The water there subsided gradually.

* 6, But there is no good unmixed withevil or no evil unmixed with good.
Although the earthquake has destroyed and damaged corn-fields either by
depositing sands over them or causing them to sink down with yawning gaps
made through them, yet it has rendered some compensation by causing some
very old bheels to be dried up, which in the ordinary course would have never
come under cultivation within the quarter of a century. These bheels are being

ploughed up and cultivated this very year, and are about to prove an apple of
discord to the neighbouring zemindars,

1 That is thatched bambco huts ¿nd brick built buildings—R. D, O.

( 22 )

NARRATIVE ACCOUNTS. 24

«7, In some places, both on the northern as well as on the eastern banks of
the river Brahmaputra, coal, burnt sands and other substances came out of the
earth along with sands and water. The samples which I have collected are sent
herewith in a separate box.

**8. Several persons réported to me that they have seen smoke come out of
the ground in a certain locality on thé eastern bank of the river Brahmaputra
opposite Begoonbary, which is on the northern. bank, and that the place from
which the smoke came out appeared to be very hot.

Fulpur, Gaira Hat- “9. The names of some of the places, where the

pagla, Barapagla, Bhug- ; ?
ly, Char Niamat, Mali- changes described above took place are noted in the

danga, Shirtha, Dhamna, margin,
etc., etc.

Houses and BUILDINGS.

“ ro. Houses were buried, but were seldom levelled to the ground. The
buildings in most places were demolished, causing immense loss to the owners,
In the town of Mymensingh all the best two-storied buildings were destroyed and
also many of one story. There were not, however, so much fissures in the town
as in other places. The town next in importance to Mymensingh is Muktagacha.
It isin complete ruin. Nearly all the buildings have been demolished, and the few
that have not been levelled to the ground have been so badly cracked, both
in the foundation and the walls, that they are no longer fit for habitation and are
likely to come down any day. The fateof Muktagacha was also shared by the
villages called Ramgopalpur, Gouripur, Kalipur, Durgapur, Atarabari and some
others which are the seats of the zemindars of this district. These have literally
been rendered houseless and have been dwelling in wretched huts temporarily
made for the purpose. Many there are who have been literally ruined owing to
all their valuables having been smashed, damaged, or destroyed by the fall of the
buildings. There are also some who will never be able to build any more. The
losses of the zemindars have been very great.

* 1I. Although the earthquake has not been happily attended with much loss
of life owing to its having taken place during the day, which is certainly a mani-
festation ‘of Divine favour to mankind, yet it has caused much suffering amongst
all classes. The upper classes I may say have suffered the most, and the lower
classes the least. |

* 12. This partial wreck of nature was for some time very awful tolook at and
distressing to ponder upon. Such an earthquake with such dire results is not
within the recollection of the oldest man living. There was an earthquake in
this district in 1292 B.S.,' some old buildings fell down and some were cracked,
but there was no change on the surface of the ground. There was also another
earthquake in this district so far back as 1254 B. S^ Its effect was chiefly felt in
the village named Poyari, where there were several fissures of small dimensions.

* I3. According to Hindu sastras, such unnatural visitations will frequently

1 A. D. 1885. 2 A. D. 1846.
e ME EE

24 OLDHAM: GREAT EARTHQUAKE OF 1807.

appear in this latter end of the Kaliyuga or Iron age. What have already ap-
peared are mere preludes. The real ones are yet in store.

“ 14. The earthquake is commonly calculated by the Hindus according to a
formula quoted below :—

“ Ifyou get famine, drought and plague in one and the same year,
you get the earthquake that year.

This calculation has indeed been verified.

“15. Apart from the physical causes which make the geological changes in
the earth, the Hindus believe in a supernatural agency, which excites the physical
causes that make the change. That supernatural agency is, again, put to action
by the vice or virtue of mankind. Changes causing woe and calamity are the
result of the former, while those causing happiness are the result of the latter.
This is what the venerable Garga says about the causes which produce un-
natural changes in the earth, the middle and the upper sky :—

* Unnatural visitations of nature are caused by too much cove-
tousness, untruthfulness in conduct and action, atheism,
impiety and impurity in thought and action. Where there is
constant repetition of the above, the gods turn their back to
mankind and cause unnatural changes in the earth, in the
middle sky, or the region of clouds and lightning, and in the
upper sky, or the region of the sun or the solar system, for
their destruction.

“ We Hindus fully believe in this that the sin of mankind has been causing all
these woeful changes, and that unless there is an advance made towards beatitude
there will be no end of these. i

* 16. As remedial measures for the mischief done by the earthquake, I beg to
suggest that the local bodies, viz., the District Boards and the Municipalities, may
be directedito co-operate with the people in opening up the rivulets or canals which
have been silted up, in re-digging the tanks that have been dried up and filled in
with sands, in removing the deposits of sands from the corn-fields, in filling up the
gaps and hollows which have been made in them, and in repairing the roads
which have been rendered impassable. It will be a matter of consideration for
the District and Local Boards whether the new canals opened up by the earth-
quake should be closed or continued to any natural stream which may be close
by. I should prefer the latter course, as they will, if continued, water a large
tract of the country and supply drinking water to men and cattle. As tothe
measures to pe adopted to guard againstthe recurrence of such unnatural changes
in future, I should only repeat what I have said in some cf my previous reports
to Government, viz., firstly, that Government should see that every man in every
community should follow the dictum of thereligion which has been accepted by
that community and act accordingly.

* 17. I am humbly of opinion that the present system of religious toleration
is not suited to India. Any man in any community should not be allowed to
violate with impunity the ordinances of the religion which has been accepted by
that community. If any man wishes to renounce his old faith and to embrace a

( 124)

NARRATIVE ACCOUNTS. 26

new one, he may do so. But then he must act according to the rules of the re:
ligion which he has newly embraced. It would, I think, be an abuse of toleration
if any man in any community were allowed to exercise his free will in acting or
not according to the religious tenets of the community to which he belongs. This
state of things brings on anomally in society and impurity in action.

* 18, And secondly, Jagna should be made according to the rule of tle Vedas.
And in order to enable one to celebrate the Jagna it is necessary that the State
should make some assignment in land to the learned aad pious Brahmans in
selected localities, who will perform the Jagna as often as the assets of the land so
assigned will allow them to do. As Hindus, we firmly believe that the calamities
which have been threatening India will be completely warded off by this means.

** rg. I earnestly solicit that our good and kind Government will be pleased to

give such consideration to my suggestions as it may think proper.”

The town of Kuch Bihar was within the area where the earth-
quake reached its greatest destructive power and probably not far
removed from the region of the seismic vertical. From this place
we have the account, written by Mr. D. R. Lyall, C.S.I., Superintend-
entof the State, tothe Chief Secretary to the Government of Bengal,

on 28th June 1897:—

“ 3. In a matter like this, personal experience is generally more valuable than
generalities, so I record my personal experience of the earthquake itself. I had
gone to the Palace and had looked on at the crowd and seen that all was going
on smoothly and then left at about a quarter to five, going out by the northern
gate, the main or eastern gate being closed for that day for carriages. I drove
round by the outside of the Palace and turned eastward infront of the main
entrance to go to my house, when, after going about 200 yards, I became aware
of a great noise in the corrugated iron sheds at the side of the road and almost
at the same timesaw a large two-storied house about 50 yards aheadof me
subside and crumble away. I pulled up and now became forthe first time aware
of the shaking of the earth. I had not felt it while driving, and [ see from the
papers others have had the same experience. As I pulled up a large fissure
opened in the road just in front of my horse, a second under my buggy and a third
close behind me, and others were visible at further distances. The road between
where I was and the Palace main gate had been nearly clear of people but at
once the crowd poured out, shouting that the Palace was falling, and I accordingly
turned back and found ٤ا١‎ Palace more or less in ruins, but the Maharajah and
his family and household were all safe with one exception. This man was struck
badly by the falling bricks and was taken to the Military Hospital where he died.
Several more sepoys and servants were slightly injured, but there was only this one
fatal casein the Palace. Upto this time I had heard no sound of the earthquake,
but as ] took leave of the Maharajah to go and see how my own house and
family had fared. there was a rumbling noise from the earth, and sand ard water
rose up from many of the fissures, In one place I saw it certainly rising two feet.

(729°)

26 OLDHAM: GREAT EARTHQUAKE OF 1897.

This was after the shocks had ceased, but in other places notably all round my
house, the sand and water appear to have come out with the first shock. It certain] y
did not do so in the Palace compound. I then drove to my house, nearly two
miles off and foundthe road full of fissures and cracks and every pucka house 1
passed more or less injured. The worst collapse was the “ Chottakuti," the fall
of which I have already described, This house being on the line of road taken by
the ** Tazzias,” a number of people were in it looking at the tamasha, and five
persons were killed by its fall. The whole loss of life was eight persons, one
patient in the hospital being killed by the*fall of the roof and the firewatchman
on the Jubilee tower having fallen withit. This man lived for twodays in spite
of the height from which he fell. These two with the five in the * Chottakuti,"
and the one in the Palace, were the only fatal cases, and wonderfully few persons
were injured. This is due to the offices all being closed and to the population
being to so large an extent out in the open air, I arrived at my own house to find
it absolutely in ruins and all round it water and sand which had been vomited
from the ground. The fissures here were much larger than near the Palace.”

Outside the area of the seismic vertical the accounts of the earth-
quake all agree in representing the motion as largely an undulatory
one, and inmany cases the presence of visible earth-waves is re-
corded. Beginning with the Assam Valley I may quote the account of
Mr. A, E. Shuttleworth, District Superintendent of Police, who was
in camp in the Mangaldai Subdivision at the time. He writes, under
date 16th August, in an official report to the Deputy Commissioner
of Darrang as follows : — | i

“On the 12th June 1897 my wife and I were sitting on the verandah of the
Chatgari resthouse waiting for a slight shower of rain to stop. At 5-13 P.M.
we were suddenly startled by a very vivid flash of lightning followed by a
tremendous crack of thunder. At the same time the bungalow began to tremble
slightly. This I at first put down to the thunder, but as the trembling motion
began to increase I cried out that it was an earthquake. The motion began
getting more and more violent, and as the timbers all began to crack and the
verandah floor to split under our feet I hurried my wife outside into the
rain, which was coming down in torrents. It was as much as I could do to hold
my wife up. We then saw the earth all round heaving in a most frightful man-
ner. The earth resembled waves coming from opposite directions and meeting
in a great heap and then falling back ; each time the waves seemed to fall back
the ground opened slightly, and each time they met, water and sand were thrown
up to a height of about 18inches orso. This shock lasted for about three minutes
I should think. There was no rumble with it. Theshock was strong enough to
knock over a couple of elephants I had in camp with me. My horse too in the
stable was knocked off hislegs and my dogs could not stand up in the verandah
of the bungalow. On re-entering the bungalow we found the verandah floor

( 26 )

NARRATIVE ACCOUNTS. 27

was one mass of cracks, the roof beganto leak badly all over, and the whole
bungalow had settled down, sothat we could not shat any of the doors that were
opened nor open the doors that were shut.”

Captain P. R. T. Gurdon, Deputy Commissioner of Kamrup, who
was in camp at Nalbari—a place which may have been within the
area of the seismic vertical, and was certainly close to it,—reports
that earth-waves were distinctly visible from the rest house compound,
which is situated close to a large pathar or series of rice-fields. The
waves could be seen following one another at intervals, the ahu rice
falling and rising as the waves progressed. On subsequent enquiry
Captain Gurdon stated that he could not recollect the length of these
waves, but that their rate of travel,though decidedly faster than a
man could walk, was not so fast as he could run. We may take it that
the rate must have been between two and four feet per second,

From Kohima, in the Angami Naga Hills, an account by a corre-
spondent to the Englishman Newspaper, may be reprinted :—

* Notwithstanding a delay which was inevitable, an account of the earthquake
in this part of Upper Assam may prove not uninteresting. First, as to Kohima,
which is a little known Civil and Military station 4,080 feet above sea-level,
standing in the midst of the wild Naga range on the Eastern Frontier of Upper
Assam, distant 92 miles north from Manipur and 123 miles south from Nigriting
(or Shikarighat) steamer ghat, on the Brahmaputra. It is in the same latitude
with, and some 200 miles east of, Shillong.

“On the 12th instant at about 5-20 P.M., local time, we felt the first shock of
the earthquake, whose violence made us somehow feel anxious in particular for
Shillong and Calcutta, where we knew the numerous masonry buildings would
have but little chance if the earth-waves were even half as severe as we had
experienced. I was sitting in my garden, facing east, just outside my bungalow
examining a flower-bed, when I suddenly began to feel dizzy from an apparent, at
first gentle, fore and aft movement. Not having been well, I naturally conceived
I was going oft into a faint and instinctively pressed my forehead and eyes with my
hands; but the peculiarity of the motion and vibration of the ground, both of which
became more violent, immediately acquainted me with the fact that it must be an
earthquake, and this was at once confirmed by hearing the creaking of the wood
and iron work of my bungalow behind and observing the consternation and
alarm of my sister who then rushed out of the verandah, and of all the servants
who also very speedily vacated the house. On rising and looking around I dis-
covered the whole house bodily rocking— I should say at least six inches each way
from the perpendicular at the highest point (25 feet)—distinctly east and west (not
to be too particular), the exact situation width-wise of the house being south-east
and north-west, and my sister staggering about trying to maintain her equili-
brium, almost believing that the world was coming to'an end ; while the cook in
horrified alarm rushed forth from his kitchen repeating aloud the Kulma and

(7221)

28 OLDHAM: GREAT EARTHQUAKE OF 1897.

calling out that there was something wrong with the bungalow roof. I doubt
if he felt re-assured when ۱ told him it was not the roof but the world that was
shaking. The other servants could not keep their feet but fell flat to the ground,
while our puppy kept running about, watching the house and barking furiously.

** [ suddenly recollected that certain valuable ornaments on the drawing-room
mantelpiece would be inevitably destroyed, and tried to make a rush to save them
but was taken aback to discover 1 could not at first move, and staggered violently
east and west as if on board a boat ina heavy swell. Eventually as the height
of the wave diminished I forced my way into the house, dodging the orchids
hanging inthe verandah which were plunging violently to and fro (east and
west) in a most extraordinary manner about two feet from the perpendicular
‘each way. Fortunately rot a single ornament had as yet suffered, and I had time
to remove them all to the ground. Thetwo mantelpieces of the house faced back to
back, north-west and south-east, that is, in the direction opposite to the wave, as
did most of the shelves and brackets, and consequently, with the exception of one
small vase laid prostrate, not a single article was thrown down during the whole
period of the wave. I afterwards discovered slight cracks extending upwards in
both chimneys, and the clock on one of them, whose pendulums swung north-
east and south-west, stopped short at 5-20 P.M., local time. There were no rum-
blings or explosion sounds. The appreciable part of the quake lasted some five
minutes as ascertained afterwards from the Telegraph Master, who, working
before his clock, had at once noted the hour, vzz., from 5-24 to 5-29 P.M., local
time, or 16-24 to 16-70 (7.e,, 4-24 to 4-30 P.M.) Madras time, which is 57 minutes
behind Kohima time.

* After the shock had passed we had time to re-enter the verandah (which I
had again vacated) and sit down and pass some remarks about the visitation,
when, some four or five minutes later I guess, we heard several loud explosions
like guns in the direction of the fort three-quarter of a mile north-west which I
remarked must be either the mountain battery practising or bombs going off in
honour of the Mohurrum.”

Southwards of Kohima, Major 1. O’Donnell, D,S.O., writes from
Kunjukphul in Manipur :—!

* The earthquake gave no warning of its coming. It began suddenly and
severely, and lasted I should say some four minutes. At first the motion was an
up and down one, and one I had never before experienced. This by degrees gave
way to the ordinary side to side kind of motion, which was at first a long and
slow motion, gradually becoming shorter and quicker till it suddenly ceased. The
hill immediately behind I could distinctly see moving, and appearedto move, turn
and turn about with the knoll we were on. At the first shock we ran all outside,
and for most of the time it lasted it was difficult to stand. I saw the others in
motion as you see a person on deck in rough weather.”

From Manipur itself Captain H. W. G. Cole, Political Agent,
writes :—! 8 ُ
! Assam Government Official Report, Appendix XI.
( 28 )

NARRATIVE ACCOUNTS. 29

“ The first tremor was brought to my notice by a servant, but after he had
pointed it out it was some seconds before I noticed any motion. The motion
then became rapidly more severe, and I went out on the lawn. By the time that
I reached it the motion was sufficiently severe to render standing very difficult, if
not impossible. The ground seemed to undulate from south to north. I momen-
tarily expected the cracking of the straining timbers to be followed by a crash of
their breaking, but just as severe damage seemed inevitable the undulating
motions decreased in severity, and were succeeded by what seemed to me a tilting
motion from north to south." *

Turning now to the south, the earthquake was observed at Silchar
by Revd. O. O. Williams, D.D., who had just arrived from Karimganj,
and was at the time in the mission bungalow. He writes :—

“ The first thing I noticed was a wave-like motion passing under my feet, some-
thing like the swing of a suspension bridge. Two or three waves of this kind
were felt, there was a perceptible interval between the crest of one wave and that
of the next, and I just felt the rising and falling of the ground, an earthquake
being far away from my thoughts. Direction from north to south. The second
thing [ noticed was that the canvas ceiling was thrown into undulations, the waves
travelling from north to south, the house being situated exactly facing the four
points of the compass. The progress of the waves from one end of the room to
the other could be followed by the eye, and my impression is that the length of the
ceiling was thrown into about five or six waves.

*Although more than surprised at this phenomenon my thoughts were with
the rats, which I believed were then running about distractedly on the moving
ceiling. 1 confess I was wondering how the rats could cause such waves, and
was coming to the conclusion that some much bigger animal was required, when,
in the third place, the walls claimed my attention. A peculiar shaking and sound
of knocking began, when the shaking and knocking suddenly increased in violence
and I saw the north end of the room, where I had located the knocking sound,
give one great bulge, the convexity being to the south. *Earthquake" now flashed
into my mind * *, I backed promptly intojthe verandah and got * * a few
yards from the house when the wave motion made it very difficult to stand.
My impression at this moment was very far from the simple one at the beginning.
It was the feeling of a boat tossed in a choppy sea, i.e., a crossing of great waves.
These waves were certainly not from north to south, it is quite possible that the

most pronounced were from the north-west."
Dr. Williams describes his movements and impressions during
the earthquake and continues :—

* | have mentioned the above details that you may form a conception how long
. it lasted. [t seemed five or six minutes; but half an hour or so after the shock

( 29 )

30 OLDHAM: GREAT EARTHQUAKE OF 1897.

when the place was cleared out a little and we were able to sit down and talk over
things, I took out my watch, went over my observations and thoughts leisurely,
allowing sufficient time for everything, and had finished the recital in 2} minutes.
I do not believe that from beginning to end the earthquake could have exceeded
three minutes by any possibility.

“I have only to add that during the great earthquake I had only the sensation
of undulating, not of shock.”

No noise was heard by Dr. Williams himself, but others in Silchar
noticed a sound, like that of a traction engine, just before the shock,
and one person, who was seated in a chair at the time, noticed the
birds rising suddenly from the trees before he himself felt anything.
This was probably due to their greater sensibility to the small preli-
minary tremors not felt by human beings,

In the accounts from Chittagong, besides clear evidence of an
undulatory movement of the ground, we find a feature repeated in
the accounts from many other places. The shock was divided into two
distinct phases, and the direction of the movement was different in
the two. Mr. J. W. Thurlow, Sub-Assistant Superintendent of Tele-
graphs, writes :—

<“ When the first shock was felt I was seated facing east, and suspected the
shock came from my left. On hurrying out I faced the north and felt a
tendency to fall forward. The motion was slight but very regular from
north to south for one-and-a-half minutes, when suddenly a peculiar quiver
was felt which made me think the direction had changed.. Acting on this sup-
position I faced the west, when I found I could scarcely keep my equilibrium,
having a tendency to fall forward. The shocks now came from west to east, and
were far greater in intensity than those from north to south. They were regular,
but few were more severe again. The severity of the shocks did not end gradually
but abruptly.”

In Calcutta, much the same was noticed in several accounts, from
among which I may select the following, communicated to the daily
papers by the Revd. Father Lafont, S.J.:—

“ The first intimation that I got of the earthquake was a kind of twisting sen-
sation and a rumbling noise. A few seconds soon convinced me that it was an

cg)

NARRATIVE ACCOUNTS. 31

earthquake, and I ran out on the terrace overlooking the portico of our College:
time about 5 P.M. Everything around was in motion, but my attention was
chiefly attracted to the energetic oscillations of a heavy stone cross, capping the
front of the College. It was clearly moving, and the whole portico as well, almost
due north and south ; a little statue in my room, placed on a bracket attached on
the north wall, fell and rested slanting against the wall. These two facts point
evidently to a first and strong oscillation north and south coming apparently from
the south.

“ Finding the terrace on which we stood unsafe I went down a spiral staircase,
and had to prop myself in doing so forfear of losing my balance. Once down-
stairs in the open I looked again at the cross, but it was then still, the oscillation
had been twisted to the east : I found in my laboratory afterwards two clear indica-
tions of that twist in two or three heavy objects which had pivoted from south to
east. In my opinion the strongest and most lasting vibration was that from east
to west, as evidenced by a tall book-case placed against the east wall in my room,
being found to have fallen direct towards the west, when I returned to my room.
Besides, all the cracks in the College buildings, in St. Thomas’ Church, and the
Loretto Convent show unmistakable signs of being the result of an east-westerly
vibration.

“I have a conclusive proof that the last vibration was in that direction, in a
maxim barometer suspended on gimbals, found to be oscillating due east and
west, as was alsoa heavy pendulum inone of my glass cases. Several heavy
objects had also been thrown down in the same direction. Thetime this unex-
ampled earthquake lasted is variously estimated from 4 to 6 or over 7 minutes. It
may be interesting to know that a slight but distinct shock was felt yesterday night
at about 9-30 and another one at about 8-30 A.m.this morning. The barometric
curve of yesterday is almost a straight line, showing an almost stationary state of
the atmosphere.” |

M. Paulde Bure, Assistant Agent of the Messageries Mari-
times, writes that he was wakened by a succession of very rapid
vibrations which were vertical as it seemed to him; then came some
slight oscillations from north to south, and it was only then that a
noise as of a heavily laden cart was heard. These oscillations in-
creased till pieces of plaster and masonry began falling down, and he
went outside the house. From here he saw that the house was
oscillating in a south-east to north-west direction, and the oscillations
were unequal as it always seemed as if the house would fall to the
west rather than to the east, A lightning conductor rod was set in
motion and swung some 20? on either side of the perpendicular, in a
. direction parallel to Park Street.

ec و‎

32 OLDHAM: GREAT EARTHQUAKE OF 1897.

From an account of the earthquake received from Dr, Pyari Lal
Mukhopadhyaya, I extract the following passage, which, it will be
seen, agrees with the Revd. Father Lafont's account in discriminat-
ing two distinct phases of the shock :—

“The first shocks or vibratory motions were indiscriminate or all sided. No
particular direction could be marked. They were rapid, say some seven or eight
per second. Then followed the undulatory motions as the rolling of a mighty
ship. They were regular, keeping even time like the movement of a pendulum,
telling the second asexactly. The motions first were from north to south as we
ordinarily call it ; but the exact direction was from north-east to south-west of the
compass. I could make it out by a hanging lantern and a cage of a canary
bird. Both of them began to swing from north-east to south-west. The rope to
whick the lantern was hung measures 8 feet 5 inches. When the shocks were
severe it swung about 35 feet on either side, that is described an arc of about 7 feet.
As these shocks seemed about to subside, some rude and, fearful jerks were felt, the
lantern and the cage moving from north-west to south-east, so the direction of
those jerks must have been from due west to east, or the lantern and the cage
could not have moved in the direction (north-west to south-east) that they did.”

As an instance of how difficult itis to tell the direction of move-
ment of the wave particle from mere ‘sensations or the movements
impressed on objects of suchcomposite character as houses, I may
quote from Mr, Hayden’s note :—

“The cracks may be divided into two sets, those running north and south and
those running east and west ; the former would be due to the fact that the structures
vibrated east and west, and the latter to vibrations at right angles to that direction ;
and during the earthquake, the existence of these vibrations wasclearly seen. A
good example is found in No. 13 Theatre Road, which house I was carefully
watching throughout almost the whole duration of the earthquake. When I reached
the compound, a few seconds after the first tremor, 1 found that the house was
violently vibrating east-west, and soon a crack opened through the house from
north to south, and the western third of the house could be seen vibrating outwards .
the large crack opening as much as 4 inches and then closing again. This was
observed from the southern side of the house. Suddenly, however, the motion
appeared to change, the crack in the south wall ceased to open, and on passing
round to the western end of the house, the building appeared now to be swinging
north to south, and vertical cracks began to open over windows and along lines
of weakness in the west wall, but they did not gapeto nearly the same extent as
the large crack in the south wall. ”

It will be seen that this account places the relative order of the
north-south and the east-west vibrations exactly the reverse of the
( 35 4

NARRATIVE ACCCUNTS. 33

Other accounts. The fact is that buildings constructed like those of
Calcutta can only swing in two directions, and the local varia-
tions in the nature of the earthquake motion, which are now suffi-
ciently well established, combined with small variations in the orienta.
tion of different houses, would be sufficient to account for either of
these taking precedence of the other in different localities.

The nature of the motion at Calcutta appears to have been ex-
clusively undulatory with a period of between one and two seconds,
Though destructive to buildings and distinctly felt by persons walk-
ing or standing on the ground, it was quite unnoticed by those who
were driving at the time, their first intimation of anything unusual
being the sight of cracking buildings and of people rushing from their
houses, unmindful of their attire in their anxiety to reach a place of
safety. The undulation of the ground, though apparently unnoticed
by horses in motion, was very distinctly noticed by bicyclists, who
were unable to preserve their balance and found themselves com-
pelled to dismount from their machines.

At Saugor Island, the experience was similar to that at Calcutta,
According to Mr. R. Gilbert, Telegraph Master, an unusual disturb-
ance of the sea preceded the first vibration, the direction of which
was from south-south-west to north-north-east, for about 1 minute
and 30 seconds ; this was followed by a stilliiess for about 5 seconds,
after which a reaction set in from south-east to north-west, lasting
with extreme severity for 4 minutes, and diminishiug in force gra-
dually till it stopped 55 seconds later.

The swaying of the lighthouse from zenith to north-west was
noticed to be greater than in the opposite direction.

From the neighbourhood of Panchkura, inthe Midnapur District,
I have received an important account by Mr, S. M. I. Williams
Assistant Engineer on the Midnapur-Calcutta Railway, which is
interesting as giving some idea of the extent of the undula-

‘tion of the ground, and of the way in which these surface waves

(33)
D

e PER

34 OLDHAM: GREAT EARTHQUAKE OF 1897.

had lengthened out from the comparatively short and conspicuous

` waves of the region within and near the seismic vertical :—

* [ was taking levels on the railway bank about a mile south of the village
of Panchkura, and was about to take a reading when I noticed the bubble
oscillating, and 1 at first thought some one had touched the legs. In 5 or 10
seconds I could feel the bank shaking to-and froin a direction which gave me
the impression of being from south-south-east to north-north-west. Unfortunate-
ly I had no watch with me. I tried to get the exact direction of the waves by
turning the level round on its axis, but the legs were not in the right position.
Stacks of bricks near me, about 3 feet high, were shaking, but did not fall.
The shock appeared to last 3 or 4 minutes, but, for at least 5 minutes after all
shaking had apparently ceased, my level indicated that the vibrations were still
continuing, and at this point I put away the level. Towards the close of the
worst part of the shock, I turned the telescope to the staff being held about 150
feet due east of it, and although the bubble was oscillating most erratically, the
level and the staff were perfectly motionless relatively to each other. When
I got back, it was 4-44 P.M. by Madras time, and I estimated it was about 15 or
20 minutes after the earthquake. I also set up the level to test roughly at what
angle it had been tilted, and found that the maximum angle during the great
shock was o? 48' and during the lesser oscillation afterwards o? 1! ; the former
angle is probably inaccurate.”

From Midnapur, Mr, Davies, Telegraph Master, reports that
the shock commenced witha rumbling, and the cot on which he
was lying rattled to the speed of it, an up and down roll such as
could be made witha signalling key, equalling a ten or twelve word
per minute pace. This roll lasted about 2 minutes, and asit passed
eastwards the doors rattled to the Azck of it and the motion changed
to an undulating south-north one. He triedtorush from the house
but had to hold on to a pillar and felt distinctly sea-sickish; the
ground was distinctly billowy, and it seemed that “ these were waves
of motion going ferward north but receding south on the forward
movement, such as a water wave at the time of tide does.” The
total duration was Ak minutes.

From Balasor the Telegraph Master reports that at the first
signs of the shock, after noting the time, he went to the battery
cells and found the liquid in them undulating from east to west;
this motion lasted for about go seconds, when the movement changed
to north to south for about the same time.

( 34) |

NARRATIVE ACCOUNTS. 35

At the same place, Conductor J. W. Turner reports that he timed
the earthquake as lasting four minutes, The motion was a long rolling
one, such as would be felt if on the deck of a ship in a fairly heavy
sea. The circuit bungalow was cracked, and a piece of the parapet
which runs east and west, was observed to swing southwards along
a crack to the extent of about g inches. The southerly movement
is said to have taken 9 or ro seconds, and the closing again more
rapidly, perhaps 5 or 6 seconds. This was repeated some seven or
eight times.

This slow opening and closing of the crack corresponds with the
information I have received elsewhere, but it must not be regarded
as the period of the earthquake wave ; it more probably represents
the natural period of swing of the moving portion of the building.

These two last-mentioned places, it will be seen, experienced an
undulation whose direction was about east-west in the first place
and subsequently in a north-south direction, whereas at Calcutta
and Saugor Island the reverse is said to have been the case.

Passing southwards down the coast, the undulatory nature of the
movement was noticed everywhere. Almost every account notices
the gentle nature of the movement, which was generally likened to
the easy movement of a ship in a gentle sea, and this movement
everywhere produced a sensation of nausea in all who are liable to
that affliction in the circumstances to which the earthquake motion
was likened,

At Bezwada, Mr. C. Napier, who describes the earthquake as giv-
ing rise to a smooth, slight rocking sensation, with slight giddiness
and the faintest sensation of nausea, noticed that certain files of
papers hanging from the edge of a table oscillated slowly from north
to south for about two minutes, The rate of oscillation is said
to have been slow, the period being about once in two seconds,
certainly not so quick as once a second. |

Returning to the north, we find the same undulatory movement
D 2 | (35)

36 OLDHAM: GREAT EARTHQUAKE OF 1897.

recorded throughout the Gangetic alluvial plain and the northern
part of the peninsular rock area. From Giridih, Dr. J. A. Dyer
writes :

“ On turning to look at the house I found extraordinary movements taking
place inthe verandah roof supported by pillars, which movements I could not
detect at the east and west ends, where there are rooms with outside walls. ‘They
(movements) appeared to be of a wavy kind, from north to south or south to north
and moved a few tiles from out their places up near the top of the rows and shot
them downwards, by what looked like small repeated shoves, to the distance of
5 tiles length in two cases, and 6 tiles in one case.

* [ could see most distinctly the heaving motion, and compared it in my mind
to what might have been produced by an elephant under such a roof, if he had
rubbed his back against it from north to south. I ought to state that my bungalow
stands east and west, and has a pucca roof over four centre rooms running east
and west and a sloping thatched roof with country tiles over the thatch, all round .
the four sides, which tiled roof is supported partly on side dressing and bathroom
wails, and front and back on strong double pillars, with beams on them, to
support the rafters. It was only over the pillars and beamsthat the movements
could be detected by me. I carefully looked tothe parts supported by the walls
of the rooms on the north side (I was on that side only during the earthqake),
but was urable to observe movements there, We stood solong in the sunshine,
that I dreaded a sunstroke, and ran for shelter and shade to a small house from
which I observed the continued movements, and 1 also observed the roof of a
godown, near at hand, vibrating upwards and downwards only."

At Lucknow, the earthquake is said to have commenced with a
series of rapid waves like a choppy sea, which gave way to longer
ones resembling a heavy swell. The former are said to have lasted
two minutes, the longer waves for one minute. The tall column of
the Martiniére swayed backwards and forwards on either side quite
three feet. |

From Allahabad, I am indebted to Mr. R. Warneford for some
interesting particulars. The earthquake here was of sufficient vio-
lence to displace crockery, etc., standing on tables and caused some
cracks to form in one or two buildings, apparently old and not of
sound construction, At Mr. Warneford’s house he noticed that
hanging lamps and saddles suspended from the ceiling were set
oscillating in an east-south-east to west-north-west direction; water
in tubs was oscillating backwards and forwards in the same direc-

( 36 )

NARRATIVE ACCOUNTS. 37

tion, rising about 1 inch on each side of the tub, about 2 feet
6 inches in diameter. Quarter of a mile away Mr. Aslett, also an
Engineer in the employ of the East India Railway Company, noticed
a series of easily distinguishable waves crossing his compound from
south south-west to north north-east. The crests were about 30 to
50 feet apart.

Further west, the reports from more than one place mention
that no shock was felt, but that an agitation of the water in wells and
tanks was observed. |

From Sehor in Bhopal, I have an account by Mr. H. de St.

Dalmas, of the Friends Mission, who writes :—

* We were not conscious of the shock by any personal sensation, and the only:
thing observable wasthe violent shaking of a door in the north wall of the bunga-
low at the north-east corner, noticed by myself, and a similar vibration of a heavy
almirah in the adjoining room, noticed by my wife. We simultaneously called to
one another to notice the phenomenon, which we could attribute to nothing else
but an earthquake. The vibration continued probably for about two minutes.
I may remark that the door, which was closed, and the almirah were both in the
same line and in the north-east part of the house, the door facing north. *

Further west, the shock was felt by the Thakur of Piploda (12 miles
west of Jaora Railway Station) in the upperstory of his house. He
was sitting at the time, and his body oscillated from side to side, while
chandeliers hanging from the ceiling swung in the same direction.

At Khandwa, the shock does not seem to have been actually felt
by any one, but a hanging lamp was observed to oscillate.

Something similar appears to. have been noticed in Bangalore by
the Dewan of Mysore, who has communicated. his account through
Dr. J. Evans; from this it appears that at the time the Dewan was
engaged ina private interview on matters of State; and that he
noticed the door moving backwards and forwards as far as the latch
would permit, just as if some person were at the door. He went
to the door and opened it, but found noone near, The exact time
was unfortunately not noticed, but it was about that of the earth-
quake, and it is reasonable to attribute the unusual and otherwise
unaccountable movement of the door to this.

At Bombay, also well outside the area within which the earth-

i )37(

38 OLDHAM : GREAT EARTHQUAKE OF 1897.

quake could be felt, Mr. Moos, Director of the Government Observ-
atory, Bombay, was fortunately engaged on the diurnal readings of
the magnetic instruments, and observed the declination needle to
swing slowly to and fro. No special estimate was made at the time
of the extent or rate, but Mr. Moos considers that the lateral
movement cannot have been far from 7 of an inch. As the length
of the suspension is about 12 inches, this gives an angular oscil-
lation of about 30’ of arc, an estimate which is doubtless a maximum
value. S

On the north we have the account of Colanel Wylie, C.S.I,,
Resident in Nepal, who writes :—

* The shock was felt distinctly at Katmandu at about 4-35 p.m. It lasted
some five minutes, and was of a double nature, Ze, the first portion of the shock
was apparently from northto south. There was then a pause of a few seconds,
and the next motion seemed to be from east to west. It was sufficiently violent
to cause a feeling of nausea and faintness, making: men sit down on the ground,
and to induce nearly every one to quit their houses. Inthe Residency grounds,
are many tall pine trees, and these oscillated violently as if they were reeds; I
have never seen anything of the kind before. At the same time, I could dis-
tinctly hear the roar of terror rising from the city, which is about a mile off. ”

The Nepalese Representative at Lhassa reported to his durbar
that a severe shock of earthquake was felt there at 5-25 P.M. on the
12th June, that it came from the south and lasted about 8 minutes.
In this report no mention is made of any damage to buildings, but
Mr. H, G. Hobson, Commissioner of Customs at Yatung, specially
mentions that no damage was done at Lhassa.

Yatung itself seems to have felt the shock pretty severely, as.
may be judged from Mr. Hobson's report,

“ Little damage was done at Yatung, where the few houses are of wood, with,
end walls of rough stone ceménted together with mud. Two ofthese end walls,
fell outward, Ze, to the N. and E., others were cracked.

* [ndoors, articles such as bottles, lamps, etc., standing on shelves, fell over to,
the N.-E. A pendulum clock, on a bracket facing about N.-N.-E.-and S.-S.-W.
did not stop.

“ A flag pole was noticed to, oscillate violently in an almost N.-E. to S.-W.
direction.

“There were minor shocks during the night of the 12th and on several days,
following, but they caused no damage.
" aequ sustained in the neighbourhood :—
. At Langrang, situated halfway between this and the top of the Jalep
( 38 )

uw

NARRATIVE ACCOUNTS. 39

pass, the Dak bungalow, a stone structure with roof of wood shingles, collapsed
entirely, most of the débris falling to the eastward.

* 2. Below Yatung, the barrier wall built of stone and roofed over the parapet
with wood shingtes, was thrown down in part and otherwise shattered, the eastern
end coming to most grief.

* 3. Overa mile below, the Ka Govi monastery, situated on a spur over-
looking the vale of the Mo Chhu, was wrecked. The main buildings fell, the
high compound walls, of concrete were cracked from top to bottom, whilst those
at the back fell to the N. and E.

"A small fort just below hadits parapetsthrown down, the wreckage falling

both inwards and outwards, say to the N.-E. and S.-W.

4. In the Chumbi valley all official residences, oarracks and private dwell-
ings were badly damaged. Compound and other walls fell mostly to the E. and
S.E. From Ki Bim Ka to Gopa Jong in the upper Tomo district. some 7o
houses were destroyed, a child being killed and four adults seriously injured by
falling walls.

* From To Mi to Do Chaka in the lower Tomo region, two houses were wrecked
and others badly cracked.

“The earthquake would appear to have pretty well expended itself in this
direction at the base of the high range of mountains to the north and east,
at the foot of which the Mo Chhu runs eastward towards Bhutan.”

“Little damage is reported from Phuri to the N. of this, and none from
Shigatze or Lhasa. In recent letters from the baier places it is mentioned that
an earthquake had been felt, but no alarm had been caused thereby.”

In Burma the motion was of the same character, 7. e. in the
main undulatory. The accounts are proportionately less numerous
and Jess full than from Upper India, but two may be selected for
quotation. The first of these is that of Mr, A. E. English, Assistant
Commissioner, who was at the time in camp on the banks of the
Theingale, some 7 miles south-west of Kyauko village and about
19 miles east of Tagaung. Between 5 and 6 P.M., he

" Noticed the water in the tank, which was an old river course, containing
about 300 yards of shallow water, lapping up against the bank below my tent.
My hunters said it must be elephants bathing, but on looking there were none at
the other end. Some one then pointed to the trees shaking, and we knew it must

be an earthquake, « s zs. The shocks were not felt, and, except for the lapping
of the water, would not have been noticed.” .

A somewhat similar account was received from Thayetmyo, where
Mr. E. Gabbett, Executive Engineer, writes that—

“I, Lieutenant Marsden, XIII Madras Infantry, and Lieutenant Sharpe, 1st
East Lancashire Regiment, were fishing in a small tank, 4o yards square (very

( 39)

40 OLDHAM ; GREAT EARTHQUAKE OF 1897.

approximately), on that evening. About 5-30 o'clock (this is only a guess, as no
one looked at their watches) the water in the tank started going backwards and
forwards just as if the water in a basin had been tilted backwards and forwards ;
the maximum rise of the water on the east end of the tank was 18 inches. This
movement lasted about 3 minutes. The movement appeared to be from west to
east as near as we could judge. People standing onthe bank could feel nothing.
No cracks, that I am aware of, occurred in any buildings or bridges.”

Some other accounts from Burma make specific mention of the
same undulatory motion. Mr. C. F, Gilbert, Executive Engineer,
Bhamo, in the course of his account of the earthquake as felt at
Nahakaung, the junction of the Katha branch of the Mu Valley line,
writes :— I have experienced considerably more severe earthquakes
in the Himalayas, but the sensation of sea-sickness which accom-
panied this one, and which seems to have been noticed by everyone
almost, is quite a new feature.” At Bhamo Mr. P, D. C, Pereira,
Inspector of Police, reports that the shock was not the usual short
jerks, but resembled more a gently rocking motion. Ft caused a
sensation of nausea in several people.

These two accounts are valuable as being comparative and show-
ing the difference between the nature of the disturbance produced
by a very severe earthquake at a distance from its centre, and the
possibly more severe, disturbance of a slighter earthquake at a short
distance from its centre.

Such are the accounts of the nature of the shock. Besides
those printed, many more have been received, but they do not add
anything of importance. From them it may be gathered that there
was everywhere a marked undulatory movement of the surface of
the ground, the waves being shorter in the central area and length-
ening out as they progressed, In the central area, and wherever
they were small enough to be seen as wave-like undulations of the
surface, it is certain that they can have only been locally produced
waves, the result of the earthquake, but not spreading far, or travel-
ling at anything like the speed of the real earthquake wave, this
being of too great a length and travelling too rapidly to be seen. At
greater distances, where the shock was only slightly felt or not felt

( 40)

NARRATIVE ACCOUNTS. 4

at all, and only rendered itself noticeable by disturbances of level of
pools of water, it is probable that what passed was a real wave of
translation, and was in fact the earthquake wave.

In the central area the horizontal to and fro movement was
large, and abrupt and very sensible, but as the wave travelled, this
became a simple undulatory movement, very different, according to
the accounts, from the more common short and quick to and tro
movement experienced in the case of less severe earthquakes which
do not spread far from their origin,

Though somewhat apart from the scope of this chapter, the effect
of the earthquake on the mud volcanoes of Kyauk Pyu may be no-
ticed. It is contained ina letter from the Deputy Commissioners
Kyauk Pyu, to the Commissioner of Arakan, No, A dated Kyauk

1—97!
Pyu, the 25th August 1897 :

* The mud volcano in this island is well known to all people, and is occasionally
active.

* On this occasion loud reports were heard coming from this volcano followed
by a flow of mud, which continued for an hour and a half.

* About 11 o'clock that night loud reports were again heard, and a new volcano
opened out 2,500 feet to the south of the old volcano. There was a very large flow
of mud. from this new crater, so large in fact that it spread out over the land near,
destroying acre 1'05 of paddy land belonging to a cultivator named Na-Ban-
San. The flow continued until about midday on the 13th June. The reports
made by the opening ofthe n ew crater were followed by a very brilliant meteor
which appeared to travel from the south to the north.

“On the 23rd June about 7 P.M., a slight shock of earthquake was felt and
another meteor was seen, and three sounds as of the distant booming of a gun
were heard.”

* This is ascertained by Maung Po from personal enquiry. The latter part is
vouched for by him as he himself saw and heard all,”

The only other reports which mention any effect on hot springs
are those of the Collectors of Chittagong and Patna. The former
reported that the burning springs of Sitakund were unusually lively
after the earthquake, while the latter reported that the hot springs
: at Rajgir (Rajagriha) were said to have discharged discoloured water
for three days.

(41)

09 0001

42 OLDHAM: GREAT EARTHQUAKE OF 1897.

CHAPTER 111.—THE ISOSEISMIC LINES AND AREA OVER
WHICH THE SHOCK WAS FELT.
The determination of the isoseismic lines, that is to say, lines of

equal violence of shock, is the necessary preliminary to determining
the varying rate of diminution of seismic energy in different
directions. Strictly speaking there should be little difficulty in deter-
mining these lines, as the violence of the shock may be regarded as
a function of the amplitude and period of the wave,'or more simply,
of the maximum rate of acceleration of the wave particle. All
these elements are capable of determination from the records of a
complete seismograph, and, in the absence of these instruments, a
tolerably complete substitute may be found, at any rate when an
earthquake is severe, in the overturning of simple objects, such as
pillars, in countries where they are sufficiently numerous,

In the present case, however, objects suitable for giving seis-
mometric results of any degree of accuracy are rare, and, owing tothe
extent of the country over which the earthquake was more or less
destructive, they have not all been examined by competent obser-
vers. It has consequently been necessary to fall back on the general
accounts of the earthquake and its effects, and to draw the isoseis-
mic lines on the basis of the effects of the earthquake on masonry
buildings and on human sensations.

In the account of the Charleston Earthquake, Captain Dutton
made use of the Rossi-Forel scale of intensities and drew
isoseismic lines, or isoseists, representing ten different degrees of
intensity. Ina closely populated and civilised country, where most.
of the buildings are of brick or stone, this was possible, but the area
over which this earthquake was felt is largely a wild, thinly populated
country, and even in the thickly populated parts brick and stone build-
ings are rare and widely scattered. As a result it has been found
impossible to attempt to define more than seven degrees of intensity
lying within their isoseists, which may be defined as follows :—

1. The first isoseist includes all places where the destruction of
brick and stone buildings was practically universal.

(42)

150551575 AND SEISMIC AREA. 43

2. The second, those places where damage to masonry or brick
buildings was universal, often serious, amounting in some cases to
destruction,

3. The third, those places where the earthquake was violent
enough to damage all or nearly all brick buildings.

4. The fourth, those places where the earthquake was universally
felt, severe enough to disturb furniture and loose objects, but not
severe enough to cause damage, except in a few instances, to brick
huildings. |

s. The fifth, those places where the earthquake was smart
enough to be generally noticed, but not severe enough to cause any
damage.

6. The sixth, all those places where the earthquake was only
noticed by a small proportion of people who happened to be sensi-
tive, and being seated or lying down were favourably situated for
observing it.

This scale does not pretend to scientific accuracy, nor are the
differences between the successive isoseists by any means uniform;
all that can be said in its favour is that it is the best which the
materials available allow of being used. As compared with the
Rossi-Forel scale it may be said that, roughly speaking, the isoseist

number
No. 1 includes degree 1o of the Rossi-Forel scale
33 2 33 3) 9 2) H
33 3 وو‎ LE 8 ر27‎ HI
2 4 ” دو‎ 6 and 7 2) 92
» 5 Tg »4 and 5 3 »
” 6 ” em. BG 3 m nm

To those who were in the area affected by the earthquake a
better idea of the scale may be given by the quotation of specific
instances of each degree. The first isoseist includes all such
places as Shillong and Goalpara, where destruction was universal ;
the isoseist runs near Rangpur and Kuch Bihar. No, 2 would
include Murshidabad, Malda, Darjiling ; No. 3, Bhagalpur, Krish-
nagar and Calcutta. Beyond this, No. 4 would include western

( 43 )

44 OLDHAM: GREAT EARTHQUAKE OF 1897.

Bihar and the eastern half of the North-West Provinces. About
Allahabad comes the isoseist No. 4 and about Agra No. 5.

It will probably surprise many who were in Calcutta at the time
to see it stand so low in the scale as I have indicated above, and
shown on the map No. 1, and when they recollect the alarm
caused at the time and the ruined aspect of the city shortly after
the earthquake the surprise is not unnatural. There are, however,
special causes in the mode of construction of the Calcutta houses
which account for the great destruction resulting from the earth-
quake.

Figure 3 represents the transverse section. of a typical Calcutta
house. There is a central portion built with thick walls, with flat
cemented roof and cemented or marble floors to each story ; on the
south is a broad verandah, where the roof and floors are carried by

YA

CALL CMM ee? 27727. Ud 22 22 > ZZ 7 7 SS /

2/7
Y I
7 291000 7 ۳ SS

Ges M "e" یسب و "کے کر‎ 2777“ ES

l well
i dÉ 1 Ge

|^ umm ill WM ln UUM A f MUM MU um MLL: a TA
WU

IIA

ot Y III M zi

Fig. 3. Typical Section of a Calcutta House..

tall brick pillars, the upper half of the openings being filled by
wooden venetian blinds; on the north is a large porch, often with a
room built over it, as shown in the figure, but sometimes without

( 44)

150581575 AND SEISMIC AREA; 45

this. The two storied houses are built in the same manner, with the
omission of the top story.

From this description it will be seen that the house is divided by
two vertical planes into three very distinct portions of very different
weights and proportions, Consequently when the house is set 2
ing each portion will try to swing as an inverted pendulum, with
its own proper period of oscillation. As this will be different in each
case, and as the difference is greatest in the case of the main body
of the building and the south verandah, the different portions, unless
strongly tied together, will alternately separate and press upon each
other,

This separation of the verandah from the main body of the
building was observed by more than one person, and the rate
at which the separation and return took place, estimated at from
7 to 20 seconds by different observers, shows that it was not
directly due to the bending of the house on the top of the earth wave,
whose period was much quicker, but to a swing communicated to the
building, and the elastic oscillation of its different parts, Tall
factory chimneys are known to sway several feet in a high wind,
and bricks and mortar have sufficient elasticity to allow of a separa-
tion of several inches between two parts of a tall house swinging
in Opposite directions.

After the earthquake it was found that by far the greater amount
of the serious damage done was of the character described, and few
were the houses, even those otherwise little injured, where two
cracks did not open as shown in fig. 3. Moreover, it was found
that where the main beams supporting the roof and floors ran
north and south, in the same direction as those of the verandah, and
hence acted in some degree as ties, the damage was much less than
when the beams in the main building ran east and west and those
in the verandah north and south.

Another cause of the damage in Calcutta was the heavy and weak
balustrades and cornices found on most of the older houses, These
will be noticed in the views reproduced on Plates XXIII and XXIV,

(45)

ہے اھ ای یں

46 OLDHAM: GREAT EARTHQUAKE OF 1897;

and are represented in section on figures 4 arid 5, fot which Í am
indebted to the courtesy of Messrs. Macintosh, Burn and Co. Figure 4
represents in section the roof, cornice and balustrade of an ordinary
private house; fig. 5, that of the Calcutta Town Hall, which may be
regarded as an extreme case of this type of construction. The balus-
trade, whose general appearance can be seen on Plates XXIII and

i

RH

مہ

N

a >
CD

Fig. 4. Section of roof, cornice, and para- Fig. 5. Section of roof; cornice, and
pet of a private house, Calcutta. parapet of Calcutta Town Hall.
Scale 4 inch = 1 foot.

XXIV has a heavy upper rail, made of brick and mortar, heavily plas-
tered over, and supported on a row of thin earthenware hollow sup-
ports, sometimes filled with mortar, but more often left empty. These
earthenware supports are generally plastered over to match the rest
of the building, but the plastering adds little to their strength, and
when the earthquake came they were in very many cases unable to
support the strain thrown on them by the inertia of the heavy upper
rail, broke across their narrowest part, and allowed the rail to be

(46)

150881575 AND SEISMIC AREA. 47

precipitated to the ground, carrying with it any projections that
stood in its way (Pl. XXIII, fig. 1).

Little less dangerous were the heavy cornices which adorn not only
the outside of these houses, but are often repeated in the interior.
They project two feet or more (in the case of the Town Hall over five
feet) and, being constructed of ordinary brick and mortar covered with
stucco, have little transverse strength at their junction with the
main wall, In the interior of the house they do not project so far,
but are often a foot or more wide and made principally of plaster.
These heavy cornices proved especially liable to damage; long
lengths of them were broken off, and in their fall caused wreckage
and damage. |

Yet another reason for the great apparent damage done was the
almost universal practice of coating the brickwork with a thick coat
of stucco. This is done for climatic reasons, a house so treated
being much cooler than one in which bricks are left bare, but the
plaster is easily detached when the whole house is caused to rock,
as was the case in the earthquake, and an appearance of great
damage was produced where, but for the stucco coating, the house
would have been almost uninjured, Plate XXIII, fig. 2,1s a view which
may be taken as typical of a large number of houses after the earth-
quake, where it will be seen that, apart from the detachment of
plaster and the damage done by its fall the house is practically
uninjured.

After making allowance for these special considerations, and for
the fact that brick-built houses are far more numerous than any-
where else in Bengal, and that, consequently, the number seriously
injured would naturally be greater, it will be found that the destruc-
tiveness of the earthquake was, proportionately, much less than
throughout the north-eastern portion of Bengal; a conclusion which
is borne out by the comparatively small amount of damage done in
the native quarters of the town, where a different mode of construc-
tion is adopted and brick buildings are, as a rule, smaller and more
homogeneous than those in the European quarters, Another source

(47)

EESTI Ta 2... Lei 0

48 OLDHAM : GREAT EARTHQUAKE OF 1897.

of strength is to be found in the buildings being, asarule, not detach-
ed but built in rows, and where this was the case it was seldom that
any but the end houses suffered, except in the case of those which
were old and badly built,

In accounts of great earthquakes, at any rate those written in
modern times, it is usual to attempt a detailed delineation of a
greater or less number of isoseismal lines, and much information can
doubtless be derived from the evidence their irregularities gives of
the irregular distribution of the energy of the earthquake. In no
case do the isoseists have that regular circular or elliptical form
which would be found if the wave was propagated through a homog-
eneous medium, but in the case of the present earthquake it has been
found impossible to draw these irregularities with any degree of
accuracy, and the isoseists actually drawn on map No. 1 are
purely diagramatic and represent their probable course if we
could suppose local conditions everywhere uniform, and the diminu-
tion of the violence of the shock comparable to that which can be
recognised in a westerly direction along the plains of Upper India.

It will be noticed that these curves come closer together on the
east and south-east than on the west and south-west, and I believe
that this closing in is, at least partly, real, and not due to imperfect
information. On the north nothing is known of the course of the
isoseists, and it is not impossible that the closing in of the inner
isoseismic lines on this side is incorrect. I have drawn them in
accordance with what little is known and with what might reason-
ably be expected, but no value can be attached to this portion of
the curves.

So sketchy a delineation of the isoseismic lines may be regarded
as unsatisfactory, and it certainly compares unfavourably with the
elaborate delineation of the actual courses of a much larger number
of lines given by Captain Dutton in his account of the Charles-
ton Earthquake of 1886, The difference is largely inevitable
but partly deliberate. Ata very early stage of the investigation of

(48) | :

ISOSEISTS AND SEISMIC AREA. 49

this earthquake it was seen that quite one-third of the area over
which it was sensible lay in regions from which it was hopeless to
attempt to obtain any information, while more than half of the re”
maining area was sparsely inhabited by ignorant and illiterate tribes,
with but few and widely separated centres from which an intelligent
account could be hoped for. In these circumstances it would have
been impossible to obtain such detailed information as would alone
have allowed of the delineation of the actual course of the isoseismic
lines. Moreover, any attempt to have obtained this information would
undoubtedly have occupied much time, besides being almost fore-
doomed to failure, so the attempt was deliberately abandoned in
order that attention might be given to those points with regard to
which this earthquake seemed likely to add to our knowledge.

This being so, I shall not enter on a detailed review of the
accounts from various places (sufficient will be found in other parts
of this report,) and will merely notice certain peculiarities in
the distribution of the energy of the shock which can be detected in
the accounts.

Two partial exceptions may also be made to the general impossi-
bility of drawing the isoseismic lines, The first is in the case of the
area over which serious damage to brick-built houses was common,
This corresponds to the isoseist between degrees 3 and 4, as defined
on p. 43 or between 7 and 8 of the Rossi-Forel scale. The other
is the outer limit of the sensible shock, and may be taken as the limit
between degrees 1 and 2 of the Rossi-Forel scale. The areas included
in these two lines are indicated on Map No, I by a darker and a
lighter tint, respectively.

The irregularities in the course of the inner line need little com-
ment; they are closely connected with the geological structure of the
country, and are largely, if not entirely, due to the well-known fact
that earthquakes are more destructive in alluvial ground than on rock
at an equal distance from the centre.

At the outer limit there are some peculiarities which need more

E | ( 49 )

ae و‎ oe

رھ

0 OLDHAM: GREAT EARTHQUAKE OF 1897.

detailed notice and the most prominent of these is the detached area
about Baroda and Ahmadabad. The Collector of Ahmadabad reports
that he did not experience the shock himself, and that though it was
reported to have been felt, he does not think that one person in
a thousand noticed anything. Even this small proportion is, how-
ever, probably larger than that which noticed the shock over a large
area to the east, and the number of separate accounts which mention
the shock as having been felt in the Ahmadabad district and Baroda
territory, though accompanied by a larger number which report that
nothing was felt, leave no doubt that the shock was just sensible on
the alluvial tract bordering the western edge of the peninsular rock
area.

Further east we find the same thing. In the alluvial area of the
Narbada Valley the shock seems to have been pretty generally
felt, though only by a portion of the inhabitants, and at the western
end it seems to have been noticeably more feeble than at the eastern.
North of the Narbada alluvium the shock was much less noticeable.
The accounts received from Sehore in Bhopal and from Piploda have
been given on page 37. At Nimach the shock was not felt by anyone
in cantonments, but the Subah, whose house is situated on high
ground overlooking the city, some z miles from cantonments, reports
that the shock was felt by him and caused the door-chains to clank
and the tiles to rattlé on the roof. The Station Master at Raoti, 17
miles west of Ratlam, reports that the shock was felt there, but all
other places, to the west of those just mentioned, report that it was
not felt; the only exception being Mount Abu where the shock
appears to have been felt by one or two persons,

One more instance of the local appearance of the earthquake as a
sensible shock must be noticed inthis western region. 1 have a
complete series of reports from the the Station Masters On the line
from Jabalpur to Bombay. From Jabalpur to Seoni they all, with a few
exceptions, report the shock as having been felt ; from Seoni, which is
still on the Narbada alluvium, to Bombay, every Station Master reports

(ya)

> Ma یو ری‎ Be. Lat: be
3 ا‎

150551575 AND SEISMIC AREA. 51

that the shock was not felt, with the single exception of Burhanpur

where it is reported to have been felt. It is possible that this excep-

tion may be due to a greater sensitiveness or keenness of observation

on the part of the Station Master, but it is noteworthy that Burhanpur

is on the northern edge of the Tapti alluvium, but close to the boun-
. dary between it and the great spread of Deccan trap.

Before leaving the western area it may be of interest to note that
in Lalitpur in Bhandelkhand, far beyond the region where any serious
injury to masonry buildings is reported, the Birdha inspection-house on
the Cawnpur-Jhansi road is reported to have suffered serious
damage. The building was of stone, with the roof supported on
stone slates carried by stone beams. A few cracks were opened in the
roof of the main building, parallel to the stone beams supporting it,
and the walls and roof ofan outhouse fell down. From the account
submitted to me it does not appear whether there was any local cause
to explain this damage which, though slight, was much greater than
any other in the North-West Provinces or Oudh, or whether it was
entirely due to age or defective construction of the buildings,

Turning to the eastern limit of the known area, the shock is said
to have been severe on the northern frontier of Burma, and to have
thrown down trees in the Hukong Valley and at Sampawn but at
Myitkhina it is said to have been so slight as not to have been gene-
rally noticed at the time. In the North Shan Hills it was felt more or
less over the whole charge, except at the Tawmo outpost east of the
Salwin.

In Southern Burma the shock appears to have been very noticeable
all down the Irawadi Valley, as all stations report that it was
felt, even as far south as Diamond Island, but it was not felt at the
Alguada reef lighthouse. To the east of the Pegu Yoma the shock
appears to have been very much less felt, and the difference cannot
be attributed to a greater distance from the centre. At Toungoo it is
reported to have been very slight, and a few only of the Sta:ion
Masters On the Rangoon-Mandalay Railway felt it. It was felt in

E 2 (su. )

777 :7ھ
d MUSS‏

52 OLDHAM: GREAT EARTHQUAKE OF 1897.

Western Karenni, and seems to havebeen felt in parts of Myelat,
though there is an uncertainty about the date, but was not felt
elsewhere in the Southern Shan Hills.

These records give an approximation to the eastern limit of the
area over which the shock was felt, but east of Bhamo we have no
records. An enquiry instituted through Her Britannic Majesty's
Envoy Extraordinary and Minister Plenipotentiary at the Court of
China failed to elicit any information as to the earthquake having
been felt in China ; but as the region where it could have been felt is
one of the most remote from the capital, largely inhabited by primitive
tribes, and as the earthquake must have been too feeble to attract
general attention, no great value can be attached to the negative `
result of the enquiry.

On the north the only information available is that contained in the
reports of Colonel Wylie and Mr, Hobson quoted in the last chapter.

Summarising the results reviewed’ above we find that the area over
which the shock is known to have been felt amounted to not less than
4,200,000 square miles (statute), or 3,120,000 square kilométers and
this does not include the detached area near Ahmedabad or any part
of the Bay of Bengal, nor does it include the large area in Thibet or
Western China, over which the shock was certainly sensible, though
we have received no reports. If we round off the area by including
these tracts we get a total area over which the shock was sensible
as about 1,750,000 square miles, (4,550,000 sq. km.) while the area
over which it is known to have done serious damage to masonry
buildings is not less than 145000 square miles, (377,000 sq. km.,)
or, if we include the area from which no reports are procurable, about
160,000 square miles (416,000 sq. km.).

These estimates, it must clearly be understood, large as the figures
may seem, are not sensational ones. That is to say, the extremest
possible dimensions have not been taken in order to attain a large
result, but so far as possible the actual areas calculated, and if the
figures are in error at allitisin defect and not in excess,

Kies

RATE OF PROPAGATION. 53

CHAPTER IV.—THE RATE OF PROPAGATION, AND TIME

OF THE COMMENCEMENT OF THE SHOCK.
- Of all the data required in seismological investigations none are
so important and none so difficult to obtain as the exact times at
which the various phases of an eathquake were felt. This is partly
due to habitual inaccuracy in timekeeping, accuracy to a second
-or even to a minute of time being seldom required in ordinary life,
but also to the nature of the eathquake wave or series of. waves.
They do not commence abruptly and cease absolutely at determin-
able moments, but the sensible shock is preceded by minute tremors
and followed by longer and slower undulations, both of which are
insensible, but in different degrees to different people. The time, for
which the preliminary and terminal stages of the disturbance last,
increases with the distance from the centre, and it would be possible
for two careful observers, side by side, to record materially different
observations of the time of commencement and cessation of the
shock, as the movements felt by one might be quite insensible to the
other,

In the case of this earthquake I have received a large number
of time records from various sources, which may be classified as
follows :—

(1) The automatic records at Calcutta and Bombay..
(2) Reports from the Telegraph Department.

(3) Reports from Station Masters.

(4) Accounts of private individuals.

The second of these categories comprises the reports which all
telegraph offices in India and Burma were ordered to submit. It
might be supposed that the time records from these would show a
high degree of accuracy, seeing that a time signal is transmitted
daily to every office. The result has not, however, fulfilled this
. anticipation. It is not in human nature to take more trouble than
is necessary to attain the purpose desired, and in small offices where

asik

54 OLDHAM: GREAT EARTHQUAKE OF 1897.

telegrams are seldom received or despatched! no great trouble is
taken to keep accurate time. In many of these small offices it is
evident that no note was made at the time when the earthquake was
felt, and the conjunction of these two causes of error has vitiated
nearly all the returns. In a few cases the time reported is obviously
in error to the extent of several hours,

From the larger and busier offices, where a careful record of time
is essential to the proper conduct of the business, the returns are of
greater accuracy and value. In these the signallers have to keep a
log-book in which the times of receipt and despatch of messages and
the alteration of the electric circuits are recorded; and where the
work is sufficient to keep the office continuously employed, it is
obvious that a considerable degree of accuracy in timekeeping is
necessary. But a high degree of accuracy, according to the standard
of ordinary life, may be a high degree of inaccuracy where even
fractions of a second should be taken into consideration, and it is
obvious from the returns that, even where accuracy is most to be
expected, the time is often in error by several minutes.

On the railway the same considerations affect the accuracy of the
returns, but here the constant traverse of trains, each of which has
to keep its running sheet in agreement with the times kept by the
Station Masters, introduces a certain check on the time kept at road-
side stations. Yet it is evident from the returns that the times re-
corded at large and busy stations, where many trains stop, are, on the
whole, more accurate than at the roadside stations where only a few
trains stop. It is also noticeable that on the main lines a more accu-
rate record of time is kept than on the branch lines, or on lines which
do not carry much through traffic. On the East Indian Railway, the
busiest, and by common consent best managed, line in India, it has
been possible to adopt a method of treatment of the records which
gives a more accurate result than could be obtained by using the

11 have myself had occasion to despatch a telegram from an office where I was
informed that mine was the first message received or despatched for two months; but
this is an extreme case.

( 54 )

RATE OF PROPAGATION. 55

recorded times from individual stations. The records obtained from
the main line stations between Calcutta and Delhi were plotted and
a curve drawn (see Plate XXXIX) in which the abscisse represent
the. distance of the stations, the ordinates the times; the curve so
obtained is irregular, but by smoothing it a curve was obtained which
gives the probable time of commencement of the shock at each
station with greater accuracy than the actually recorded times.

The same method was attempted in the case of the Eastern Bengal
State Railway line from Calcutta to Darjiling, but in this case the
individual variations in time between neighbouring stations, and the
very large proportion which gave what was evidently merely the
nearest five minute interval renders the method inapplicable, In this
case, as in that of all the other lines from which returns have been
obtained, it has been necessary to adopt the less satisfactory
method of selecting those of the returns which appear on the face of
them to be most trustworthy.

The fourth category, voluntary reports from private individuals,
varies most largely in value, In a certain number of cases every
care seems to have been taken to ensure accuracy, and it is expressly
stated that the watch or clock, by which the time was taken, had
been or was subsequently compared with the timepiece of the near-
est station or telegraph office, or with the daily gun. All these stan-
dards are, however, liable to errors which may—for the present
purpose—wholly vitiate the result.

A fruitful source of error, and one that it is often impossible to
eliminate, is the different times that are kept and used. I do not refer
to isolated localities where there is neither railway station nor tele-
graph office, and where the local time is determined by the gastric
sensations of the individual in charge of the station time-gong, or
the indications of a sundial, more often than not constructed for a
different latitude to that in which it is set up, and probably adjusted to
the meridian with no more accurate instrument than a cheap pocket
compass ; but to those places where the presence of telegraphic
communication should enable accurate time to be kept.

Ki, Së

56 OLDHAM: GREAT EARTHQUAKE OF 1897.

The official standard time, used by the telegraph department and
by all railways in India, is that of the Madras Observatory, 5 hours,
20 minutes, 59'2 seconds}, east of Greenwich. But the Indian
telegraph system ranges over nearly two and a quarter hours of longi-
tude,? and, as the hour-zone system has not yet been adopted, the use
of Madras time would be inconvenient in many cases. A system has
consequently sprung up of using local time, this being obtained by a
correction of an integral number of minutes, the number to be added `
or subtracted being printed in the official telegraph guide.

This, in itself, would introduce no element of confusion if the
system was uniformly adopted, but itis not. Throughout the Madras
Presidency, Madras time is used, except at Vizagapatam, where local
time is used. Elsewhere local time is that in general use, but where
the difference is not great, railway time—as that of the Madras
meridian is usually called—is used in some places, local time in
others. Yet another source of confusion is to be found in the practice
of using a local time which is not that of the locality, but that of the
nearest large city ; for instance, at many places in Bengal, the time
conventionally used is that of Calcutta, not that of the meridian of
the place or standard time.

As a consequence of this divergence of time-systems in use it is
necessary to be certain, even after an observation has been accepted
as on the face of it good, which system has been used. Asa rule,
there is no difficulty, for the difference between them is great, the only
possibility of confusion is at places near the meridian of Madras,
and in Bengal, when the local time may not be that of the locality,
but of Calcutta,

In the reports on great earthquakes it has been usual to print at
length the time records obtained. In the present case to do so
would take up a large amount of space and serve no useful purpose.

1 According to the most recent determination; the old value was 5 hours 20 minutes
€9'4 seconds.

2 According to the Indian Telegraph guide the most westerly open station is
Chaman, 55 minutes slow, and the most easterly Keng Tung, 75 minutes fast, of Madras.

E Se) j

RATE OF PROPAGATION. 57

It is only necessary to refer to the reports on the Charleston ! Earth-
quake, or those of Agram 2 (1880) and Laibach ? (1895), to see how
largely even what may be regarded as careful records can vary, and
those of the great earthquake of 1897 exhibit an even greater diver.
gence. I shall consequently confine myself to noticing those records
_which appear to be most accurate.

In making such a selection great care should be, and has been,
taken to exclude, as far as possible, any bias to select those records
which give a result most concordant with the anticipated rate of
transmission. Were the inherent probability of the recorded times
solely taken into consideration, or even largely so, it would be possible
to obtain ample evidence for almost any assumed rate of propagation.
The proper course is to look only to the inherent probability of the
record being correct, as judged from the circumstances of the report-
er's occupation, or the details of the precautions taken to ensure accu-
racy. In some cases records which on the face of them carry a
presumption of accuracy, are so discordant with the others of the
same class, that they must be rejected on this ground alone, but
no good can result from accepting a large number of records whose
only claim to credit is a consistency with other records, this con-
sistency being perhaps due to accident or a combination of error$
which happen to compensate each other.

Turning now to the detailed consideration of the records, we
may begin with those from Calcutta and Bombay, where they
attain a greater degree of accuracy and trustworthiness than at
any other places.

At Calcutta a tide-gauge is permanently established. The time
is checked every day at 4 P. M. and 6 P, M. and the times of the
disturbances recorded on the afternoon of 12th June can consequently

! Captain E. Dutto. The Charleston Earthquake of 31st August 1886, Ninth
Ann. Rep. U. S. Geol. Survey (1889).
2 F. Wihner, Das Erdbeben von Agram am 9th November 1880, Sitz, ber. d.
` Math. naturw. Cl. k. Akad. Wiss., Wien. LX XXVIII, pt. I, 1884.

$ F. E. Suess, Das Erdbeben von Laibach am 14th April 1895. Jahrb. kk
Geol. Reichsanstalt. XLVI (1897).

(- Sa?

38 OLDHAM: GREAT EARTHQUAKE OF 1897.

be depended on within the limits of the accuracy of measurement of
the diagram. The results of the measurement, as communicated by
Major S. G. Burrard, R.E., in charge of Tidal and Levelling operations
are as follows :— |

The first effect of the earthquake was to cause the ‘level of the water to fali
1 inch, between 4-53 P.M. and 4-55 P.M., June 12th.

At 4-55 P.M. the water rose suddenly 4 inch, after that the change in the water-
level was normal till 5 P.M., when it rose and fell 3 inches in rapid succession many
times for a period of 3 minutes. After that a just perceptible alternate rise and fall
of at most 1 inch continued till 5-21, when the change in the water-level again
became normal ; at 5-54 this slight agitation recommenced and lasted till 5-56.

The first small disturbance between 4-53 and 4-55 P.M. was
doubtless seismic, and is the only indication we have of any pre-
monitory tremor. It was not, however, a sensible disturbance, at any
rate there is no record of its having been felt by any one, and it
may be neglected for purposes of comparison. The main disturbance
commenced somewhat suddenly immediately after 5h. om. P. M.

Another instrumental record was obtained at the Alipur Observa-
tory inthe disturbance of the barograph trace, a reproduction of
which is given in fig. 6. This would have been a valuable record

Fig. 6. Portion of traceof the barograph at Alipur on 12th June 1897, showing
the effect of the earthquake.
SE )

RATE OF PROPAGATION. 59

if the time could be depended on, but I am informed by Mr, C.
Little, Meteorological Reporter tothe Government of Bengal, that
no special care is taken to keep the time strictly accurate, as an
error of one or two minutes is not material, and the time recorded
on this diagram is not to be depended on for a greater accuracy
than this. The time of the maximum disturbance is about. 5h. 3m.,
which shows that error of the clock must have been well within the
limits mentioned.!

The only detailed time observation, other than instrumental, which
has been communicated to me, is that of Babu Pyari Lal Mukho-
padyaya, of 71-1 Bencatola Street, who writes:—

am a medical practitioner residing in the native quarter of the town.‏ ا“
Having prepared myself for my usual evening round at about two minutes short‏
of five in the evening of the 12th instant, my Garrhy ready at the door, I had‏
a short chat with a gentleman waiting for me. Just as I was stepping in, I felt‏
somewhat giddy and questioned myself as to what ailed me. I thought something‏
wrong in the head and waited a moment. When lo! everything began to swing,‏
and I found my mistake out. The shocks then were slight but rapid, resembling a‏
sort of general tremor, and the directions could not be apparently marked. I‏
immediately took out my watch, which was correct to the minute. It was 5h, īm.‏
23s. The watch had been compared with the gunfire the very same afternoon.‏
In order to make assurance doubly sure, I compared it with the gun-fire the next‏
afternoon, and found it as accurate as | expected. Some 15 or 20 seconds must‏
have passed before I could collect myself, and, making allowance for the distance‏
from which I heard the booming ofthe gun, say some 12 seconds it takes the‏
sound to travel that distance, the exact time of the earthquake must have been at‏
about 5h. om. ۶‏

1 The record is a photographic one, a ray of light falling through a narrow slit on a slowly
travelling band of sensitised paper. The mercury column stands in front of this slit ; and, as
it rises and falls, protects a varying amount of the paper from theaction of the light. The
downward point of the curve during the earthquake is consequently due, not to an actual fall
of the barometer, but to a disturbance of the mercury column by which its upper surface was
caused to oscillate and admit light to a level below the mean height of the barometer. That
such movement of the mercury column did take place is shown by the report of the Meteorclo-
gical Observer at Cbandbali, to the Meteorological Reporter to the Governinent of Bengal,
in which he says “as soon as possible I looked at the barometer, but the mercury was
agitated to such an extent (pumping) that the nearest mean reading | could get was 29'650.
At 4 P.M. it showed 29:674, Ther. 83°, nearly motionless; and at 5-30 P.M. 29'700, Ther. 83°,
pumping slightly."

From the construction of the instrument, no tilting of. it as a whole could affect the
record, as the change in height of the barometer, due to such inclination as could have
been given to it, would be inappreciable.

n sp.)

60 OLDHAM : GREAT EARTHQUAKE OF 1897.

From Budge-Budge, to the south of the town of Calcutta, Mr,
Thomas Harris, Deputy Wharf Supervisor, Customs Department,
writes :—

“On Saturday, 12th instant, I left my office a few minutes to 5 o'clock and
proceeded to the petroleum depót to supervise the landing of kerosine oil cases
from the; S.S. Nedjed, at No. 3, Port Commissioner’s pontoon, and exactly at

5 o'clock by my watch (this time is daily compared with the telegraph office-
clock), which keeps excellent time, I felt the ground. beneath: me quake, followed

by a rumbling sound. ”
At the Government Central Telegraph Office the time of the

commencement of the shock was noted by Mr. Sub-Assistant Superin-
tendent Stow as ¢ minute after 5.

I have also received information, both verbal and written, from
many individuals, all of whom claimed to have noticed the time of
the commencement of the shock and determined it by watches which
had been checked by the daily time-gun at one o’clock. The times
given vary from I to 3 minutes past five, a very few being beyond.
that limit."

The times recorded agree very closely; that of the tide gauge;
which gives the commencement at 5h., of Babu Pyari Lal Mukho-
padyaya who makes it 51 seconds after 5, of Mr. Harris who makes
it immediately after 5 o'clock, and of Mr. Stow, who made it
All are in very fair agreement with other careful

4 minute after 5.
observations, and the true time may be assumed as about 5h. 1m..

P.M. by local time.

At Howrah Railway station, the terminus of the East Indian Rail-
way, the time, according to the Station Master, was 16h, 28m., Madras:
time. The Fairlie Place, Chitpore and Armenian Ghat offices of the
same company give 16h. 27m. Other stations gave returns which

are evidently only approximate.

1 Many clocks were stopped, two of these being the astronomical clocks at the Alipur-
Observatory. One, whose pendulura swung east-west, stopped at 5h. 3m. 5'31s P.M. local
time ; the other, whose pendulum swings north and south, at 5h. 4m. 11'71s. It is well known
that no dependence can be placed on the time of stoppage of clocks, as it never ccincides with
any definite phase of the shock; moreover, a pendulum may go staggering on for some

es after it;has received the impulse;which ultimately leads to the stoppage cf the clock.

(Go)

minut

RATE OF PROPAGATION. 61

The Station Master of Sealdah, the terminus of the Eastern Bengal
Railway, gives the time of commencement as 16h. 27m.

A mean of these returns would give the time as about 16h. 27'5m.
and the time as obtained by plotting the returns from the stations on
the East Indian Railway and smoothing the curve as described above,
gives about 16h. 27m. 40s. as the time of commencement at Calcutta.

Calcutta local time is conventionally, in accordance with the
official Telegraph Guide, 33 minutes 4n advance of Madras or Railway
time, but the local time actually used is that of the daily time
signal controlled by the Alipur Observatory, which is 32m. 20s. in
advance of Madras, and this is consequently the correction to
be used. In the case of Mr. Harris’ record from Budge-Budge,
as the time was obtained from the Telegraph office, the correction of
33 minutes would have to be applied. Reducing the observations to
Madras or standard time, we have—

Tide-gauge - à : - : IÓh. 27m. 40s.
P. L. Mukhopadyaya . - : . 16h. 28m. 37s.
Telegraph Central Office . . e 16h. 28m. ros.
T. 2 ‘ : , : : 16h. 27m. os;
Railway stations, mean . . e 16h. 27m. 30s.

Do. Smoothed curve. . a 16h. 27m. 4os.

All these times are liable to an error of 30s. and cannot be regard-
ed as accurate within that limit; an arithmetical mean gives 16h.
27m. 49s., which is probably as close an approximation as circum-
stances permit, With observations so few, and with so large a probable
error in each case, more refined mathematical treatment seems out of
place and would only lead to.an appearance of accuracy which has
no foundation in fact, The time deduced may be taken as accurate
within a quater of a minute, and the local time of commencement of
the shock was between 5h. and 5h, om. 30s. P.M. of local time.

At Bombay, which was beyond the limits over which the shock
could be felt, we bave the records of the magnetic instruments
detailed in chapter XI, and also the barograph trace which exhibits

,8 disturbance precisely similar to that of the Alipur barograph.

Mr. Moos, the Director of the Observatory, tells me that the clock of
| | ( 61 )

62 OLDHAM : GREAT EARTHQUAKE OF 1897.

this is regulated with the same accuracy as those of the magnetic
instruments, and that the time derived from it may be regarded as
accurate within one minute at the outside,

Abstracting the results, we have the following times of com-
mencement and cessation of movement as derived from the records
of the different instruments:—

Commencement. Close.
Declination . : e ۰ : . 16h. 5m. 16h. 34m.
Horizontal force . - O . 16h. 2.5m. 16h. 36m.
Vertical forge : M ES . 16h. 6.5m, 16h. 14m.
Barograph . : . e : . 16h. 5.5m. 16h. 12'5m.

It will be seen that the disturbance of the horizontal force
instrument commences 24 minutes before that of the declination,
while the vertical force shows no disturbance till 11 minutes
later. The disturbance of the barograph must necessarily be purely
mechanical, and the early commencement of the disturbance of the
horizontal force instrument is probably the same, but itis doubtful
which of these we should compare with the commencement of the
sensible earthquake at other places. The movements which affected
the horizontal force instrument are very likely due to the small
unfelt tremors which run in advance of the main earthquake wave
and gain on it as they travel from the origin. This being so, it is the
barograph and the vertical force instrument that more closely re-
corded the advent of the phase of the disturbance which corresponds
to the commencement of the felt earthquake at places further east.

How far the difference in the recorded times, 16h. 5'5m. and
16h. 6'5m. is due to a difference in the sensitiveness of the two
instruments, and how far to errors of the timepiece, or of reading
the traces, it is not possible to say. Probably the best result would
be obtained by taking a mean of the two, or 16h. 6m. local time.
As the difference in longitude between the Madras and Bombay
Observatories is 29m. 43'55., this makes the Madras time 16h. 35m.
43°5s., or say 16h. 35'75m., If the time of commencement is taken from
the horizontal force instrument this would be 16h, 32'25m. or. 3'5m.
earlier.

( 62 )

RATE OF PROPAGATION. 63

At Rangoon the tide-gauge shows a disturbance of the curve
commencing at about 5h. 36m. P.M. local time, and lasting till6 P.M.
As there is no Observatory at Rangoon the time is presumably taken
from the Telegraph office and subject to the correction of 64 minutes
given in the Telegraph Guide. This would make the Madras time of
commencement 16h. 32m. This record is, however, not of the value
that could be wished for, and the time of commencement of the dis-
turbance of the curve cannot be determined within two or three
minutes. It begins gradually, instead of abruptly, as in Calcutta, and
the inking in of the pencil trace having been done somewhat clum-
sily has added an uncertainty to that which would, in any case, have
attached to the determination of the exact commencement of a dis.
turbance which began gradually.

The other time observations, or such of them as have been
selected as being, primd facie, probably more accurate than the rest,
may now be recorded. When a place is mentioned as a railway
station and no further authority is given, the time is on the
authority of the Station Master; in all other cases the authority
for the time recorded is given.

Siliguri.—Eastern Bengal State Railway, terminus, and junction
with Darjeeling and Himalayan railway : commenced 16-26,ended 16-30.

Goalundo,— Eastern Bengal State Railway, terminus : commenced
16-27, ended 16-32.

Sonada (Darjiling). Mr. M. Livermore, manager of the Turzum
Tea estate, Nagri spur, writes—“I have an excellent timekeeper
which I frequently compare with the local telegraph office. The earth-
quake commenced at 4-58 P.M." This corresponds to 16h. 26m,
Madras time,

Darjiling.— Darjeeling and Himalayan railway, a terminal station:
commenced 16-26, ended 16-29.

Hugli.— East Indian Railway, a junction: commenced 16-27, ended
16-32.

i 9s.)

Eeer

64 OLDHAM: GREAT EARTHQUAKE OF 1897.

Naihati.— Eastern Bengal State Railway, a junction: commenced
4

16-28, ended 16-31.

Chittagong.—From this place I have received two reports through
the Telegraph Department which approximately fix the time at
which the earthquake reached that place. In the first of these the
Telegraph Master writes :—

“An entry by Sign. F. S. DeSouza in the Log Book of the Ca 34 instrument

reads as follows: ‘16-27 severe shock of earthquake.’ Sign. DeSouza says he
made this entry at the moment the shock occurred and then ran out.”

The second is the account of Mr. J. W. Thurlow, Sub-Assistant
Superintendent of Telegraphs :—
We all experienced a shock. This being immediately followed by a second.

I drew my watch and noted the time, viz., 5h. 13m. 4S, + æ these shocks did not

end gradually but abruptly, at 5h. 16m. 25s. * » I hurried into the office x* ٭‎
and at the same time compared my watch with the office clock, allowing 46 mts. as

difference between Madras and local time. I found I was 1 mte. slow, so that
the actualtime of occurrence of the earthquake at Chittagong was 5h. 14m. I
allowed the 4 seconds for not at once taking out my watch on the first shock, but
I really do not think I was later than 4 seconds.

Chittagong local time is 46 mirutes in advance of Madras,
according to the Telegraph Guide; 5h. 14m. P. M. corresponds, there-
fore, to 16h. 28m. Madras time. A mean of these two observations
gives 16h. 273m. as the probable time.

Khana.—East Indian Railway, a junction: commenced 16-26,

ended 16-33.
On the westerly line we have—

Saraghat.—Eastern Bengal State Railway, terminus of metre-
gauge: commenced 16-28, ended 16-283.

Damukdea.—Eastern Bengal State Railway terminus of broad-

gauge: commenced 16-27, ended 16-32.
Luckeeserai.— East Indian Railway, a junction : commenced 16-29,

ended 10-34.
Giridih.—A terminal station : commenced 16-28, ended 16-32.

Gaya.—East Indian Railway, a terminus : commenced 16-20, ended

16-33.
( 64 )

RATE OF PROPAGATION. 65

Chapra (Saran).—The telegraph master reports the shock as
commenced. 16-29% and lasting 4 minutes.

Buxar.—East Indian Railway: commenced 16-30, ended 16-34;

Benares.—Oudh and Rohilkund Railway : commenced 16-30-
ended 16-313. The Government Telegraph Master reports: com-
menced 16-293, lasted 34 minutes.

Mirzapur.—East Indian Railway : commenced 16-31, ended.16-35,

Faizabad.—Oudh and Rohilkund Railway, a. junction: com-
menced 16-30, ended 16-32.

Naini.—East Indian Railway, a junction : commenced 16-30, ended
16-341.

Allahabad.—East Indian Railway, an important station: com-
menced 16-30, ended. 16-34, This time is said to be. correct within.
one minute

The Telegraph Master reports :—Commenced 16-30, lasted 3
minutes, From the. Kuttra telegraph office in Allahabad the
return was commenced 16-33, lasted 2m. 33s.

Mr. R. Warneford writes that the mean of the times of 4
watches, which were observed at once, gave the commencement as
31% minutes past 4 P.M., the greatest difference being 2 minutes.

Cawnpur.—East Indian Railway, an important junction ` com-
menced 16-30, ended 16-32. The Government Telegraph. Master,
gives: commenced 16-30.

Mr. C. Curtis, Traffic Superintendent, Cawnpore and Achnera Rail-
way, writes :—

* [ was the only one writing at the time and first noticed the earthquake owing
to the table shaking and asked the others not to shake it. They replied they were
not doing so, and | saw that they were sitting away- from the table (not touching
it). I then looked under the table thinking it might be a dog, scratching himself,
but seeing nothing there, and the table still shaking, I said * It is an earthquake,”

the station master immediately looked up at the clock in the telegraph office right
in front of him, and said ‘16-30.’ ”

Utripura.—27 miles from Cawnpur ; the Station Master reports it as
“felt just at 16-31,when No. 7 up whistled, who is arriving here at16-32.”:
Lucknow.—Oudh and Rohilkund Railway, an important junction;
commenced 16-30, ended 16-35. The Superintendent of Telegraphs
F (65)

66 OLDHAM: GREAT EARTHQUAKE OF 1897.

reports that he noted the time of commencement of the shock as
nearly as possible 16h. 30m., the same time being noted in
the Lucknow telegraph office and in the Dilkusha office. The town=
clock in the People’s Park stopped at 16-30.

Bareilly—Oudh and Rohilkund Railway, a junction station :
commenced 16-30, ended 16-33. This is evidently a mere approxima-
tion, Mr. A, L. Bickers told me that he observed it, commenced 16-32,
ended 16-35. The Telegraph department return gives the time of
commencement as 16-33. |

Hoshangabad.—Mr. G. G. White, Executive Engineer, gives
the time of commencement, which was noted on the spot, as 4-35 P. M.,
railway time.

Gwalior.—The Telegraph Master reports commenced 16-32, ended
16-34 ; the times were noted in the log-book,

Ghaziabad.—East Indian Railway, a junction station : commenced
16-32 ended 16-35. This time is said by the Station Master to be
“correct to a second, the times were tested by the office clock imme-
diately the two shocks were felt.’

Delhi.—East Indian Railway, an important station: commenced
16-30, ended 16-33; the Telegraph Master reports commenced 16-30.

These are evidently approximations, being the nearest half-hour.

Kotah.—Captain W. Stratton, Political Agent, reports that the
shock was felt at 4-29 P.M., railway time, which is telegraphed daily
to Kotah, and was carefully taken. Itis evident, however, that this,
time is some minutes too soon.

Jaipur (Rajputana).—Mr. J. R. Irwin, Resident, writes : “ By my
watch, which آ‎ believe was at least approximately right by Railway
time, the earthquake was felt about 4-34 P. ×٠۶ The Meteorological
Observer noted the time as 16-38 by his clock, which is compared
weekly with the Railway station clock.

Anand.—Bombay, Baroda and Central India Railway : commenced
16-35, ended 16-36.

Dabhoda.—Ahmedabad and Parantij Railway : commenced 10-345

ended 16-34 125.
( 66% )

RATE OF PROPAGATION. 07

Dehegam.—Ahmedabad and Parantij Railway : commenced 16-343.
ended 16-341.

Bilimora.—Bombay, Baroda and Central India Railway: com-
menced 16-35, ended 16-354.

These four stations are the only ones in Western India from which
the returns have any pretension to accuracy. They are interesting
for comparison with the Bombay records.

To the south there are two isolated stations whose records could
not be advantageously combined with the above.

Kharagpur—53 miles from, on Sini-Midnapur Railway. Mr. C. R.
Duggan, Engineer in charge of this district, writes سب‎

“ The vibrations were distinctly felt for the first time at 16-29, Madras time, the
time being taken by my watch, which could not be more than a minute wrong, as
it is frequently checked by the Railway clock in the Telegraph office. The
movements ceased at 16 hours 31 minutes 15 seconds.”

Bezwada.—East Coast Railway : commenced 16-29, ended 1631.

The time records received from Burma are extremely imperfect.
The returns from the Station Masters not only exhibit the same
irregularity that is observable in those from the branch and smaller
lines in India, but there is besides a constant error, running through
the whole of them, which can only be explained by supposing that
there was an error, apart from that referred to below, in the time
used by the Railway administration and supposed by them to
represent Madras time.

From the returns by Telegraph Masters I select the following as
apparently good :—

Kindat : : ` 16h. 31m., Madras time,
Kalewa : : SN, Som, GE ~]
Akyab : . 16h. 30m. o
Ru 4 e 16h. 26m. »
Thayet-myo , « 5-32 P.M.— 16h. 31m, P
Alan-myo . 20533 5» e të, 32m |... 4
Henzada À 1834 C e om SE »
Le-myet-hna , 16h. 32m. 5

5)

Yandoon : : | 16h. 28m,

68 OLDHAM: GREAT EARTHQUAKE OF 1897.

All the other returns received are evidently approximations and
in many cases palpably inaccurate,

Though these times are given here as appantly good records,
it has not been possible to make any use of them, as it was found
that they were too discordant, There is, besides, a constant error
of some two or three minutes of time in the majority of them, which
is easily explicable. The daily time signal originates from Madras,
but before it reaches the Burma stations it requires to be repeated
so often by relays that a considerable retardation occurs. I have
been informed that when, some years ago, the longitude of Man-
dalay was determined telegraphically, ft was found that the ordinary
daily time signal was between two and three minutes late in arriving
there. This error, if applied to the time records from Burma, would
bring most of them into very fair accordance with those from Northern
India, but as the error exists, and as its amount is unknown, it has
not been possible to harmonise these, time records, and no further
use will be made of them.

Turning now to the discussion of these records, we have those
from Calcutta and Bombay forming a class by themselves as regards
accuracy and authoritativeness. The others, comprising those alone
which seem on the face of them good observations, fall into three
geographical groups. The first of these comprises a series of stations
running north and south, between Calcutta and Darjiling, and those -
lying within a hundred miles or so on either hand of this line.
The second forms a long series stretching approximately west.
ward across Northern India, and the third another long series,
stretching approximately south-south-east to south-east through
Burma,

In making use of these time observations to deduce the rate at
which the earthquake travelled, it is necessary to know the centre
from which it started, and hence it might be considered more appro-

priate to defer this chapter till after the discussion of the position
( 68 )

RATE OF PROPAGATION. 69

and extent of the seismic vertical in Chapter X. But the time
observations themselves have to be taken into consideration in this
discussion, and for this reason their description, and the investi-
gation of the rate of travel of the earthquake, are undertaken here,
although this course involves a certain assumption of what is only
established in subsequent chapters.

In calculating the distances of the stations at which we have
time reoords from the epicentre, it is necessary to assume this as a
point; but, as will be seen later on, this was a large area, and it
is consequently necessary to determine what point to accept as
that of departure of the earthquake wave. Two courses are open,
either to take a central point in the focal area or, as all our stations
lie to the west, one near its western limit. In either case, certain
errors are possible,

Supposing the earthquake to have been due, as suggested in
Chapter X, to the formation of 'a great thrust-plane, the movement
probably commenced near the centre and spread outwards. Now the
Tate at which the fissure spread may have been the same as, greater,
or less than, that at which the earthquake waves travelled. In the
latter case, if we assume a starting point at the western limit of the
focal area, itis evident that all the calculated distances will be less
than the true distances which the earthquake wave had travelled,
for the wave would have reached the assumed centre before the
disturbance starting from that point was originated. If, on the other
hand, the formation of the fissure started at more points than
one, or, if it spread at a greater rate than that of travel of the
earthquake wave, then all distances calculated from the centre of
the focal area would be greater than the true distance travelled
by the earthquake wave. The probability is that the rate of
spread of the fissure, and of the travel of the earthquake wave, were
practically the same; and in this case we may reckon from a
point near the western end of the focal area when deducing
‘the rate of travel of the disturbance. An assumed centre in
Lat. 25° 45’ N., Long. 9o? 15’ E., has accordingly been used

(69 )

70 OLDHAM: GREAT EARTHQUAKE OF 1897.

`

in all the following calculations ` this being more or less of a com-
promise!.

Taking this as the starting point, we find the distances, mea-
sured along the surface’, to Bombay and Calcutta are 1208'3 and 255
statute miles, respectively, while the time interval was 8 minutes.
From this we find the average rate of travel of the earthquake wave,
as between these distances, was probably 119.1 miles per minute, or
3 kilometres per second. The time interval cannot be depended on
for absolute accuracy; it might possibly be as much as one minute in
error, but, seeing the nature of the records by which it was obtained,
it is probably not more than half a minute out in excess or defect.
Assuming this as a probable error, we get the values of the rate of

propagation as — '

Per Minute. Per Second.
e EE men
Miles. Miles. Feet. Kilometres,
Maximum 126'5 2°10 11,135 32
Minimum 111'7 1°86 9,725 ENZ
Probable mean 119*1 1°98 10,4£0 3'0

Making all. allowances for possible inaccuracies, there remains
the fact that the mean rate of travel of the earthquake wave at the
sutface of the earth was about 120 miles a minute, or 3 kilometres
a second, as between Calcutta and Bombay.

The remaining observations are not sufficiently accurate, or
sufficiently concordant to permit of their being advantageously
treated by direct calculation, and a more satisfactory result is
to be obtained by the graphic treatment adopted in the hodo.
graph on Plate XXXIX, Here the two lines at right angles to
each other represent, respectively, a scale of distances and time,

1 It may be pointed out that, as the stations from which we have time observations are
in an approximately east and west line, an easterly or westerly shifting of the assumed focus
would make little or no alterationin the difference of distance of any two stations from it.
The only exception to this is in the case of stations like Calcutta, which lie far south of the
general E. W. line; and even here, when compared with distant stations, such as Be
the difference is not such as to appreciably vitiate the results obtained.

2 These distances, and all others in this memoir, are calculated toa mean radius of 3,956
mies and a spherical globe. The errors so introduced are trivial in comparison with the
inevitable ersors of the time observations,

LZ)

RATE OF PROPAGATION. ji
all the time observations reférred to above, except those from
"Burma, are plotted, the series of stations lying nearly west of the
origin being distinguished by having their names printed below,
while those which lie to the south or north-west, off the general
westerly line, have their names printed above.

Taking the westerly group we have, in order of their distance.
from the origin :—

Name of place. Distance. Time.

Sara Ghat . . > > > > . e 138 16h. 28m.
Damukdea œ . . . ٠ e $ . 140 16h. 27m.
Luckeeserai . j S e . e 1 e 260 16h, 29m»

Giridih , : : : > e » : 270 16h. 28m.
Gaya . . ۰ . e D e . . 336 16h. 20m;
Chapra . ` a - : d : : : 349 16h. 293m.

usar > $ : : à A o : e 391 16h, 30m.
Benares . è ? 5 à A 2 i 1 448 16h. 29811.
3om.

Mirzapur e : - > : $ * e 476 16h. 31m.
Faizabad . : e s 3 e 4 ۹ - 496 16h. 30m.

Naini x : , ` : ò ^ > » 518 16h. 30m.
Allahabad , è . . e Ti ` > 522 16h, 30M.»
33m.,
314m.
Cawnpur. $ ~ $ A $ $ b 613 16h. 30m.
Utripura ۰ e > . $ n . ٠ . 1030 16h. 31m.
Lucknow $ : $ ? 2 - e S 632 16h, 3om.
Bareilly . e . ° è e . > > 688 16h. 32M)
33m.

Hoshangabad . à è S ` è 3 e 725 I6h. 35m.
Gwalior . : ò pui , ^ : © 755 16h. 32M.
Ghaziabad . e e a , ` ^ 812 16h. 32m.
Delhi. e . 3 e 3 . à i 820 16h. 30m.
RS A RR AE MT ae Tuis poate | ` Ke 1۴۴
Anand . . e : e ° e e e 1107 16h. 35m,
Dabhoda ٠ S ٠ . ٠ e ٠ > 1,109 16h. 34m.
Dehegam ° ` . shiek” è . s^ ئ۹۹ ى+‎ 16h. 343m.

. Bilimora. e . e e - e . e ےے7٦‎ 16h. 35m,

The second group is less numerous and comprises—
Parbatipur . - e A e ٠ è s 7 16h. 25m.
Siliguri . e - i e à e . a 112 16h. 26m.
Goalundo è e . 3 è n e e 131 16h. 27m.
Sonada . ^ : - : - : . 4 160 16h. 26m,
Darjiling . n : b b A : $ ` 167 16h. 26m,
Hugli . e A e e e ۰ Wat 16h. 27m.
Naihati. - Be : a : . 09229 10h. “SoBe.

Khana . è : . A b R : e 27 16h, 26m.

Chittagong oa e . é e ° . e 237 16h, و.27111‎
28m.

Gidni sy : à e , è " é 340 16h, 29m.

(71)

92 OLDHAM : GREAT EARTHQUAKE OF 1897.

The third group of observations, those from Burma, have not been
utilised for reasons given on page 68,

Turning to Plate XXXIX, we see that the observations show a
fairly regular relation between the time of commencement and dis-
tance from the epicentre. These records fall so well together that it
is possible to draw a line through them, which represents the rate
of progression of the earthquake, and from which we can deduce the
rate of travel at any point with greater accuracy than is possible
by direct calculation from the times recorded.

Two such lines are in fact drawn on the plate. Of these the one:
drawn firm is that which most closely passes through the centre of-
the constellation of time observations, and represents the probable
true curve of the times of arrival of the shock at different places ;
the other, drawn as a dotted line, is that which, for reasons that will
appear in the sequel, is theoretically more probably correct, though
not so closely in accordance with the recorded observations. In either
case the meaning of the line is that, taking any point on the
line, a place at the distance from the origin indicated by the scale of
miles at the bottom would have felt the earthquake commence at
the time indicated by the scale of minutes at the right hand end
of the diagram. Taking first the gently curved firm line, and tabu-
lating the results obtained from this we find that, taking 16h, 26m-
as the time at which the earthquake wave first reached the surface
and started on its outward course, the distance travelled in—

1 minute was . e . . 5 > . > . . 170 miles.

2 minites ,, Xm PN MM C ET
3 وو‎ 33 . . D e e D ۰ D e 350 دو‎
4 وو‎ » e . ٠ ` . e e . e 460 5,
5 7 ود‎ . . . . . e D ۰ e 580 an
D وو وو‎ . . ° D e e ۰ e " ZOOM
7 35 وو‎ » ° . . . D ° ° D 840 وو‎
8 9 2 > e . e . . e . . 000. 5.
9 ap » e . e e H ۰ ۰ e I50 112;

Or, in other words, the distance travelled in the—
1st minute was |

S e . 5 . : . 170 miles,
20d. ss Ka . . . a . e e IIO .,,
3rd 33 » H . e D . D D . . 80 دو‎
4th وو‎ m e D D e D

e e D 110 ,9

(72)

RATE OF PROPAGATION. 73

sth minute was S NE CT e - . . : „ I20 miles.
6th 27 وو‎ . - $ . » 120 FT]
7th a an . e E . . e . - ٠340-9
8th وو‎ an H D . . . 150 وو‎
gth 27 وو‎ s C) ۶ ۰ D . . 160 39

From these tables it will be seen that the slowest rate of travel
seems to have been in the third minute and the third hundred miles,
On: either side of this limit, as we approach the centre or recede from
it, there is seen to be a steady and regular increase in the rate of
travel, - This is evident from the diagram if it is borne in mind that,
from the manner in which it was constructed, the rate of travel is
indicated by the slope of the line drawn, the more nearly horizontal
this is the greater the velocity indicated, and the more inclined the
slower the rate of travel. If the diagram is examined, it will be
seen that the line can be divided into two parts, one near the origin
which is concave upwards—indicating that the rate oftravel is
diminishing—and the other away from the origin, which is convex
upwards, indicating a progressive increase in the rate of travel,
The point of junction of these two parts of the hodograph is that

where the rate of travel was least and lies at about 280 miles
from the origin.

"The form of the hodograph drawn on Plate XXXIX resembles‏ ۔
that of Schmidt's hodograph!, based on the assumption of a higber‏
rate of transmission of the earthquake wave in the deeper seated than‏
in the more superficial portions of the earth's crust, and also the form‏
of the same author's modification of Seebach’s hyperbola,? when the‏
curvature of the earth's surface istaken into consideration, and this‏
might, at first sight, seem to favour its acceptance as the true one.‏
On closer examination, however, it is found that the acceleration at a‏
distance from the epicentre is too small to be accounted for by either‏
of the hypotheses taken separately, and still more so by the combin-‏
ed effects of the two causes, Ze, of the known curvature of the‏

" Jahresheft Ver. f. vaterl, Naturkunde in Wurttemberg, 1888, Ps 254.
2 [bid, 1890, p. 227.

(73)

24 OLDHAM: GREAT EARTHQUAKE OF 1897.

surface of the earth, and of the assumed increase of velocity of proe
pagation with the increase of depth of the wave+path,

Rejecting Schmidt's hypotheses as, in this case, requiring à
greater variation in the curvature of the hodograph than is indicated
by the recorded observations, we must fall back on Professor Milne’s
suggestion that the shock which was felt was not the direct result
of waves propagated from the centre through the earth, but of
waves of an undulatory nature, originated at the surface in the
epicentral tract, and propagated outwards along the surface, at a
uniform rate of travel.

lf this is so, the apparent variations in the rate of travel do not
exist, and the hodograph would bea straight line which may be
drawn through the Bombay and Calcutta times, continued to 50 miles `
from the assumed centre, and thence carried horizontally to the
origin. This, which is shown by the broken line on Plate XXXIX,
would agree with the recorded observations almost, if not quite, as
well as the cuved line, and would give a uniform rate of propagation
of about r19 miles per minute outside the epicentral tract.

So far as the time observations go, there is nothing to enable us >
to decide between the two, for, though the cutved line more closely
represents the mean of the observations plotted, yet the straight line
does not anywhere depart from it by more than about half a minute of
time, while, if we exclude the very discordant records such as those
from Allahabad and Bareilly, and give less weight to those which
give times of 30 or 35 minutes past 4 o'clock, and are consequently
merely approximate in their intention, it will be seen that the
straight line represents the mean as closely as does the curved line,
when all the observations are taken into consideration and given
approximately equal values. When, however, the records of the
unfelt earthquake at distant observatories come to be considered in
Chapter XV, we will see that the second explanation, that represent-
ed by the dotted line in Plate XXXIX, is the more probable one of
the two, and we may take it that the earthquake wave travelled

(74 ) |

el ee Pon‏ حر رپ ہیس

RATE OF PROPAGATION. 73

along the surface, at a tolerable uniform rate of about 120 miles
an hour.

A point of some importance to determine is the exact time of
origin of the shock or rather that at which it first reached the sur-
face. In doing this, we may take the records of places within the
central area, such as Shillong, and deduce the time directly ; or obtain
it by calculating back from the observations at places outside that
area. Of these two methods the latter gives the best results, at any
rate in this case, but before discussing the result so obtained,
the few observations that were made in the central area may be
noticed, ۱

At Shillong the exact time of the commencement of the shock
is given by Mr. J. C. Arbuthnott, C.1.E., Deputy Commissioner, as
5-11 Or 5-12 P.M.; the Assistant Superintendent of Telegraphs,
Shillong division, in his official report says—“ As soon as the shaking
was over, I looked at my watch, and allowing for the error of
my watch by that day's time, found the time was 5-16 P.M. I esti-
mated the duration of the whole earthquake to be about 13 minutes.”
This would place the commencement at from 5-14 to 5-141 P.M.
Babu Nishi Kumar Ghose, Agent of the Oriental Life Office, writes
that his time-piece, which was regulated with the gun at 1 P.M., was
stopped by the earthquake at 5-13 P.M. Other clocks dug out of
the ruins are said to have shown times varying from 10 to 15 minutes
past 5. Taking 5-13 P.M. local time as a probable mean, and de-
ducting 47m., the correction given in the Telegraph Guide, we obtain
16h, 27m. as the probable Madras time of origin, with a possible varia»
tion between 16h. 25m, and 16h, 29m.

This estimate can be checked by the returns from other places
situated within the central area, or close to its borders. The Tele-
graph Master of Goalpara reports the commencement as 5h. gm.
P.M., or 16h, 27m., Madras. From other telegraph offices there are

‘either no returns or the times given are evidently mere approxima-

Dons, At Kuch Bihar, which, as will be seen in chapter X, was
| FER.

Lé m :

76 OLDHAM: GREAT EARTHQUAKE OF 1897.

close to the central area, if not within it, I have a report from Babu
Kedarnath Mozumdar, Superintendent of Works, Kuch Bihar State,
who was looking at the clock in the Lansdowne Hall at the time
and noticed that it was 2 minutes past 5 (16h. 25m., Madras time).
The Station Master of Siliguri gives the time as 16h. 26m.; of
Parbatipur as 16h. 25m.; and the same time is given by a number
of neighbouring stations on the line, being probably the nearest
even 5 minutes. All these tend to confirm the Shillong time, and
we may take it provisionally that the time of origin was not later
than about 16h, 26m, or 16h. 27m., Madras time.

This time can, however, also be obtained, with a greater degree
of accuracy, from the hodograph. It will be seen that none of the
time records from the central area, just referred to, have been incor-
porated on Plate XXXIX, the reason being, partly their divergence
and the absence of any indications that any one of them could be
regarded as of greater accuracy than the others, and partly the fact
that any error in the position of the assumed centre would intro-
duce a proportionally very large error in the deduced distance, an
error which becomes practically negligible in the case of the more
distant stations. This being so, the curve of the hodograph, for its
first 150 miles, has been filled in by inference, and drawn so as to
harmonise with the rest of the curve and with the results which have
been obtained from other earthquakes. Consequently the point at
which it cuts the central line is subject to all the risks inherent to
the method of extrapolation. The time obtained, 16h, 26m., Madras
time, may be out more or less, but I do not think that the error can
be more than one-half minute of time either way, while the error is
probably not more than 15s., and more probably of defect than excess.
The time at which the earthquake first made itself sensible at the
surface may be said to have been between 16h. 25m. 45s, and 16h.
26m. 50s., Madras time, or between 11h. 4m. 45s. and 11h, 5m. 30s.,
Greenwich mean time (civil), of the 12th June 1898, a result which
shows as close an accordance with the time directly obtained as
could be expected. |

(76)

RATE OF PROPAGATION. 77

Before leaving this subject it may be well to devote a few words
to the consideration of the degree of accuracy of the results obtained,
This I regard as much greater than might reasonably have been ex-
pected, or than I myself expected, from the data available. It is
. true that the hodograph is a one-sided one, as no observations of
any value were obtained except to the westwards, and to this extent
its value is discounted, as the more distant observations could not
be checked by corresponding ones lying in a different direction from
the centre, On the other hand, the observations in that one direc-
tion show a surprisingly close agreement among themselves, an
agreement which is in no way the result of the selection of observa-
tions on account of their concordance. All the observations, which
are given in this chapter, were selected in the first instance on the
grounds stated above (pages 63 to 67), because, from internal evi-
dence of care or other reasons, they appeared intrinsically more trust-
worthy than those which did not, on the face of them, appear to
possess that value which would alone make it worth while consider-
ing them, The selection was made before the distances were cal-
culated, and was moreover originally made at a time when it was still
believed that the earthquake originated in a very different
situation to what afterwards turned out to be the case, a difference
which would have introduced a very appreciable modification in the
relative distances of all the places lying within 300 or even more
miles of the centre finally accepted.

No subsequent additions were made to the selected list, though
there are a large number of observations which show a close agree-
ment with the times deduced from the hodograph, nor were any sub-
sequently rejected, except those mentioned in this chapter. But
briefly, it may be said every care was taken to prevent the results
being affected by any bias, conscious or unconscious, and they may, 1
consider, be accepted as accurate at any rate, within an error of less
than five per cent.

(77)

78 ` OLDHAM: GREAT EARTHQUAKE OF 1897.

CHAPTER V.—THE RATE AND RANGE OF MOTION OF
THE WAVE PARTICLE.

In determining, or more correctly speaking giving a numerical
value to, the violence of an earthquake, there are two quantities
that have been used ; one the greatest velocity or rate of movement of
the wave particle and the other the greatest acceleration or change
in the rate of movement, In the absence of self-recording
seismographs these may be approximately determined by the size
and dimensions of objects overthrown and by the distance to
which objects have been projected by the earthquake.

In the case of this earthquake the data are not so abundant
as could be wished for, owing to the paucity of objects suitable for
seismometry, and as the details of the results of the calculations
are given in Appendix C, together with a short discussion of the vali-
dity of the formule in use, they will not be repeated here. It will be
shown there that the results obtained from the formule which
give the values of the two elements, velocity and acceleration of.
wave particle, are notin accordance with each other, and the explana-
tion of this want of accordance discussed. Here it will suffice
to say that, even if we cannot compare the values obtained by two
different formula at two different places, or even at one and the
same place, we may still compare the results obtained by the same
formula from similar data and so see whether there is any differ-
ence noticeable in the deduced violence of the shock.

Taking the maximum rate of acceleration first, we have the
following probable average values :—

Dhubri : . 2,700 mm. or oft per sec. per sec.

Goalpara » e 4,200 » » 14 » وو‎ 27
Gauhati e 0ء‎ 7 » 12 an وو‎ 7
Shillong . . 3,600 35 an 12 an m T
Cherrapunji « 3,000 > AOON وی‎ m وو‎
Sylhet . 3,000 m وو 12 وو‎ 29 m
Silchar e وو 4 35 ” 0ء‎ 55 >>

RATE OF PROPAGATION. 79

In each case higher and lower accelerations have been obtained, some of
which have been rejected for reasons given in Appendix C. Others
represent real but local variations in the rate of acceleration, which
is known to vary within wide limits even in closely adjacent spots.

The values given are what appear to be fair average ones for the

whole station.

, Comparing these results it will be seen that the violence of the
shock was practically the same at Shillong, Gauhati and Sylhet,!
somewhat more at Goalpara and less at Dhubri, while at Silchar, it
was decidedly less. It is to be regretted that no numerical value
can be given for the violence of the shock at Rangpur or Kuch
Bihar.

At Cherrapunji the shock would appear from the figures to have
been slightly less violent than at Shillong, but it was quite as de-
structive, the lesser acceleration being apparently made up for by
the greater range of movement.

In the foregoing passage reference has only been made to the
acceleration in a horizontal direction, as deduced from the formula
in common use. This has been employed as it is the only one
available, though it seems certain that no great value can be attached
to the results where, as in the present case, there was a large vertical
component in the motion. At Shillong, Gauhati, and throughout the
epicentral tract, stones have been projected upwards, as described in
Chapter IX, and this fact shows that the acceleration in an upward
direction must have been greater than that of gravity, that is to say
more than 32 feet per second per second. How much greater it is
impossible to say but in places it may have been four or five
times as great.

Put into simple language this means that the measured violence

1 lt must be noticed that the violence, or power of destruction, is. not the same as the
destructiveness, or destruction actually caused, and this is especially true where, as is generally

the case, the terms are used solely with regard to the effects produced on buildings. Here many
other considerations come into consideration besides tbe violence of the shock,

(79)

So OLDHAM ; GREAT EARTHQUAKE OF 1897.

of the shock was equal to that of a sudden arrest after a fall of about
8 feet. That the effect produced on human beings and animals was.
not the same as that of such a fall is not difficult to explain; it
is due both to the smallness of the range of motion, and to the fact
that its direction was oblique and not directly upwards, thus allowing
the muscles and joints to yield and so diminish the violence of the.
shock. i

Fishes were differently situated; attacked on all sides and over
the whole surface of their body by a blow of greater violence than this,
they were killed in myriads, as by the explosion of a dynamite cart-
ridge, In the Garo Hills, even where the rivers had been unaffected:
by landslips, the fine fishing pools of the Sumesari river were found!
devoid of fish ; I was informed too by the Garos that the number of
fish caught in their traps was phenomenally small in 1897, and
that for days after the earthquake the Sameswari river was choked’
with thousands of dead fish floating down from the upper
reaches, In the Borpeta subdivision of the Kamrup district the fish:
were killed in the same manner, and I was informed by Captain:
Gurdon that on his way in from Nalbari to Gauhati he saw two float--
ing carcases of Gangetic dolphins which had been killed by the-
shock.

At Dilma, in the Garo Hills, I got some evidence to. show that the-
shock was locally of sufficient violence to disable men. I was.
informed that two Garos were drowned in one of the lakelets, formed.
by the Chedrang fault, through which a village path used to run..
The lakelet is due to the damming up of asmall drainage valley. by the
fault, which has here a throw of about 25 feet; the stream is a very:
small one and the valley open, so that it is improbable that the water. `
would accumulate sufficiently to drown men in full possession of their
faculties and able to escape, but if we suppose that the violence of
the earthquake was sufficient to stun them it is easy to understand.

how they might be drowned by the collecting waters before they
recovered consciousness,

(. 9027

MOTION OF WAVE PARTICLE 81

The way in which forest trees have been snapped across in the
neighbourhood of these faults, as described in Chapter IX, shows that
the violence of the shock must have been very great, probably four
or five times as great as at Shillong or Gauhati. If we assume that
it was only twice as great, or the acceleration at least 64 feet
per sec., it would be equivalent to the shock of sudden arrest after
a fall of 64 feet; if three times, then the corresponding fall would be
144 feet, and so on, and as the men must have been close to the line
of fault, where the focus came up to the surface, it is difficult to put
a limit to the possible violence of the shock.

This, however, is conjecture, incapable of support by numeri-
cal estimates or calculation. All that can be regarded as certain is
that violent and destructive as the shock was at such places as
Shillong, Gauhati, Goalpara, it was still more violent at other places
Within the epicentral tract, where fortunately there were no towns
or populous settlements to feel the full power of the earthquake to

destroy.

For the maximum velocity of movement, | propose to take only

the values obtained from bodies projected by the shock. These give

for—
Goalpara > e 3,300 mm. or «11 ft. per sec.
Gauhati . . 2,400 دد‎ 7 8 89^. 99 33
Rembrai a è 4,200 » T] 14 وو‎ an وو‎

Silchar . > e 450 27 33 5 وو‎ a وو‎

These figures do not come near those of some other earthquakes.
For instance, if Von Humboldt’s statements are accepted, the maximum
velocity of movement of the wave particle must have been 80 feet
per sec. in the Riobamba earthquake of 4th February 1797,' as
against 14 feet per sec., the highest recorded here.

Seeing that the maximum velocity, maximum acceleration,

( 8r )

' R, Mallet, Neapolitan Earthquake of 1857, Vol Il, p 340.

82 OLDHAM: GREAT EARTHQUAKE OF 1897.

amplitude, and period of an elastic wave, are so related to each other
that, given any two, the others can be determined, it might be thought
that from the data available we could deduce both amplitude and
period of the earthquake wave; and from the data so obtained to
form some idea of the violence which accompanied the high velocity
obtained at Rembrai, When this is attempted, however, it is found
that the results are absurd, and the only possible conclusion that can
be come to is that the high velocities of movement required to ac-
count for the way in which stones were shot through the air,
to a distance of as much as 84 feet, were not those of the earthquake
wave, but due to bodily displacements of the ground ; movements
which left the hills permanently higher than they had been before
the earthquake.!

As this means that the motion of the ground due to the earth-
quake waves was complicated by movements of a much larger scale
of a very different character, it is evident that we cannot employ the
high velocities obtained at Rembrai to form any sort of estimate of
the acceleration, or violence, of the earthquake at that place. We
are thus deprived of the only means of attaining even an approximate
estimate of the extreme violence attained by the earthquake,

and must remain content with such vague guesses as have been
attempted.

The only direct measure of the amplitude of the wave, or of the
greatest backward and forward movement of the wave particle, was
at Cherrapunji. Two oblong masonry tombs, standing close together
and oriented with their longer axes north and south, have been
partially destroyed on the inner sides, the space between them being
filled with debris. On the outer sides, they are almost intact, but the
tombs have been driven bodily down into the ground, and on either
side, to east and west, there is a depression with a vertical side
parallel to the outer surface of the tomb and a smooth flat bottom

' See Appendix C.
(82)

MOTION OF WAVE PARTICLE, 83

over which the base of the tomb has slid. The pair of tombs and
the depressions are shown in section in fig. 7.

18" 10"

۱ Fig. 7. Section of tombs in the cemetery at Cherrapunji,

The edge of the western depression has the grass growing undis-
turbed up to the edge of it, and along the edge small fragments of lime
and plaster show that this was originally in contact with the edge of
the tomb, which has now moved away to a distance of 18 inches. On
the east the edge of the depression is raised and the grass and earth
forced upwards by the thrust of the tomb against it; the breadth ot

this depression is ro inches. :
These depressions are evidently due to the backward and forward

movement of the tombs in the ground, or rather the backward and
forward movement of the ground against the tombs, which remained
comparatively stationary in consequence of their inertia. If we
assume that the two tombs behaved as a single mass, but that after the
shock they were left in a position slightly different from their original
one, the range of motion of the earth would be the mean of the
widths of the two depressions, that is 14 inches. But the earth on the
east has obviously been forced up, while that on the west has not; the
movement on the east side has consequently received a check which
that on the other has not, and we may take the wider depression, 18
inches, as representing the range of movement. Athird supposition
is that the two tombs have not exactly retained their original relative
position, but have been pressed closer together: if this was due to an
easterly movement of the western tomb, then 18 inches would repre-

sent the range of movement added to the displacement of the
E ( 83 )

84 OLDHAM: GREAT EARTHQUAKE CF 1897.

tomb, while ro inches would represent the range of. movement, less

any easterly shifting of the eastern tomb which may. have taken
place.

Whatever the interpretation may be in detail, it is clear that the
extreme range of motion cannot have been less than 10 inches, may
have been as much as 18 inches, and was probably about 14 inches;
the amplitude, or range of wave particle on either side of its original
position, being half these amounts,

Large as theamplitude appears there can be no room for doubting
its correctness, but this probably does not represent the normal ampli-
tude. The cemetery at Cherrapunji is situated near the edge of a
high sandstone scarp, and the presence of this free surface would
lead toa locally greater amplitude than what prevailed in these hills
generally. In a somewhat similar position, on the Bálpakrám plateau
in the Garo Hills district, I found vacancies separating the soil from
the larger rocks, against which it lay, that indicated quite as large
a range of motion.

At Tura I obtained some direct evidence indicating the range
of motion. The house occupied by Mr. Dring, of the American
Baptist Mission, is a wooden frame house resting on wooden posts,
which are not continued into the frame of the house. After the
earthquake, Mr. Dring noticed that some of the posts had rocked
in the ground with the movement of the house, but others had stood
firm and the framework of the house had slid over their top. The
extent of this movement, as shown by the marks on the under side
of the frame, was 5 inches away from the post, or a total range of
10 inches, Another wooden house, also belonging to the Mission
but built with posts sunk into the ground, had a wooden flight of steps
leading down to the ground; after the earthquake it was noticed that
their lower end had been thrust out about 5 inches from its original and
final position. Assuming, as is fair to do, that there had been an equal

displacement in the opposite direction, where its traces could not be
( 84 )

FISSURES AND VENTS. 85

seen, we may conclude that the house moved ro inches relatively to
the ground. In either case the relative movement was probably less
than the actual, for the houses would, to some extent, be dragged
along by their attachment, and it may be taken that the amplitude of
the wave at Tura was not less than 6 inches, or the total backward
and forward travel of the wave particle 12 inches.

I was also informed by Mr. Dring that after the great shock was
over, he went across to a neighbouring house, which had been wrecked,
and on the road he saw a fissure, the opposite sides of which were
moving up and down to the extent of a couple of inches. Besides
these measurements of the range of motion I noticed throughout
the Khasi hills, that vacant spaces by the side of large boulders
werecommon and frequently four or five inches across, even when
Isaw them, about six months after the shock; originally they must
have been larger, and it may, I think, be taken that throughout
the whole tract lying west of Shillong and Gauhati, as far as the hills
extend, and probably over a large area of the plains besides, the
amplitude of the wave motion was nowhere less than 3 inches,
while in many places it was over 6 inches, the extreme range of
motion of the wave particle being twice these amounts.

CHAPTER VI.—EARTH FISSURES, SAND VENTS, AND ALLIED
PHENOMENA.

Fissures in the ground, and the formation of vents from which
sand, water and mud have been poured out, are phenomena which have
been known to accompany almost every really great earthquake,
and even moderate ones, where favourable conditions have existed,
but in no case, of which there is historic record, have they been so
widespread or so numerous as in the great earthquake of 1897.

The theory of their formation was first clearly understood in the
case of the Cachar earthquake of 1869,! and the explanation has been

' ۰ Mallet and T. Oldham : Notice of Some Secondary Effects of the Earthquake of roth

January, 1869, in Cachar: Quart. Jour. Geol. Soc., XXVIII, 255-270 (1872). See also Mem.
Geol. Surv. Ind., XIX, 46-60. (1882).
( 85)

co

6 OLDHAM: GREAT EARTHQUAKE OF 1897.

so generally accepted and repeated in subsequent memoirs and
textbooks that it might seem unnecessary to reproduce it here.
This will, however, be done briefly, in order that the additions to our
knowledge, consequent on the effects of the last earthquake, may be
more clearly understood, and the- repetition will besides enable me
to bring out more clearly the fact that the fissures, which are formed
by, or at any rate are observed after, a great earthquake, belong to
two distinct types, a distinction which does not seem to have been
generally recognised and sufficiently emphasised before now.

One class of fissures comprises those superficial ones caused by
the disturbance of the surface soil, or occasionally of solid rock. These
fissures probably start at the surface and penetrate down to the
depth, whatever it may be, that they reach. They are a result, more
or less indirect, of the wave motion, which has been set up elsewhere
and travelled to the place where the fissures are formed. The
other class is of deep seated origin, and when a fissure of this class,
which I propose to distinguish by the name of fracture, appears at
the surface, it may be taken to have been propagated upwards from
below. So far from being a result of the wave motion, fissures of
this class must be regarded as a cause, and as surface manifestations
of the deeper seated fracture by whose formation the wave motion
was originated. In their immediate vicinity there will usually be
found indications of an abnormal violence of shock, which falls off
rapidly on either side, while in the case of fissures ofthe first class no
such local excess of violence will be noticed. Only those fissures
which fall into the first class will be noticed here, a discussion
of those belonging to the second will be found in Chapter IX.!

1 As an instance of what I believe to be a confusion of these two types of fissures, 1 may
quote Dr. J. F. Julius Schmidt's account of the Vostizzia earthquake of 26th December 1861.
(Studien über Erdbeben, 2nd ed., Leipzig, 1879, p. 68.) In the description and map (plate IV),
a great fissure is shown running along the foot of the hills for about three miles, and is said
to have been marked by a change of level of the opposite sides to the extent of 6 feet or more.
Besides this number of smaller and more irregular fissures, accompanied by sand vents are
described as occurring in the alluvium near the sea shore. Dr. Schmidt attributes all to the
same cause 2 displacement of the alluvial deposits, but after a careful study of his description,
I find it dificult to resist the conclusion that the great fissure represents part of the focus
of the earthquake, and is in fact a fault, while the others are secondary effects of the earth-
quake similar to those treated in this chapter.

( 86 )

FISSURES AND VENTS. 87

The mode in which the fissures, with which we are at present con-
cerned, are formed is fully dealt with in Vol. XXVIII of the Quarterly
Journal of the Geological Society of London, and at page 52 of
Volume XIX of these Memoirs. It will consequently be unnecessary
to go over the same ground in detail, but an explanation of the
process involved may be given in as few words as seem compatible

Fig. 8. Diagram to illustrate the formation of earth fissures.

with clearness. In fig. 8 let g g represent an upper stratum of the
alluvium, cut by a river channel, and o oa sandy or muddy bed of Rss
cohesion, along which the upper stratum g can slide over the lower s s.
When g g is thrown into movement by the earthquake wave there will
occur a time when all the particles between the lines @ and Û will be
moving towards the river, and all to the left will be moving in the
opposite direction. In these circumstances the momentum of the
mass a A which is partially free to move over the layer o o,
sets up a strain which tends to part the surface layer at æ, the
plane of separation between the portions of the surface layer
which are moving in opposite directions, that is to say, in opposite
semiphases of the wave; if this strain is sufficiently great, a fissure
will be formed and thepiece a will move forward towards the
river channel, and this movement may well give rise to fresh fissures by
the strains set up in the moving mass.

I have spoken of the fissures as entirely due to the horizontal
movement of the wave particle, and this horizontal movement would
' of itself be sufficient to account for, and is probably the main factor in.

their formation, but for the ejection of sand and water a certain amoufit
| | (87)

88 OLDHAM; GREAT EARTHQUAKE OF 1897

of vertical movement is necessary. In fig. g suppose the wave to be

Fig. 9. Diagram to illustrate the formation of sand vents.

emerging in the direction of the parallel lines in ss. The upper sur-
face of this layer will move upwards against the stratum o o, of loose
sand or mud and water, and the wave motion will be transmitted
through it to the upper layer g g. But the inertia of this will cause o 0
to be strongly compressed, and if there is an opening through to the
surface, whether pre-existent or a fissure formed by the earthquake,
this pressure may be sufficient to force sand and water to the surface.
By the repetition of the pressure a succession of spurts may be forced
up; but a large body of semiliquid matter, once set in motion, does
not come to rest at once, and so, by the momentum of the moving
water, its flow would be made more or Jess continuous, and the
spurting, which has been described as taking place, is to be
attributed to a closing in of the sides of the fissure, and consequent
forcing out of the water between its wálls, rather than to the direct
action of the earthquake wave on the underlying stratum.

After the passage of the earthquake the pressure on the soft
water bearing stratum will be relieved, and the water which had been
forced out may be sucked back with sufficient force to wash down

the upper edges of the fissure and form a crateriform hollow with
deeply scored sides.

Turning now to aconsideration of the special facts revealed by this
earthquake, which lead to a modification or addition to the explanation
given above, there is one which stands out prominently, that the

existence of a free su rface, such as that formed by the river channel
( 88 )

FISSURES AND VENTS. 89

in fig. 8, is not essential to the formation of fissures in a compressible
material, such as forms on alluvial plain, Throughout the region where
fissuring was common the fissures are certainly more numerous and
conspicuous near the edges of river channels, or of the artificial
reservoirs known in India as tanks, but they are not confined to such
positions and may be met with in positions where they cannot be
accounted for by the movement of a portion of the alluvium bodily
towards an excavation, whether natural or artificial,

The possibility of the formation of such fissures is a direct conse-
quence of the soft and unconsolidated nature of the soil, but the
mechanism is not so clear. It seems improbable that the inertia
of the stretches of alluvium thrown into movement in opposite di-
rections could cause this compression and subsequent separation
along the lines of fissure; the late Mr. R. Mallet showed that they
could not be the direct result of a wave of elastic compression! ; and
the most probable explanation is that their formation was in some way
connected with the visible undulations of the ground, induced by the
earthquake, which travelled as independent quasielastic or purely
gravitational waves. Such waves traversing the alluvium would
throw each part into alternate compression and extension as the
surface.was bent into a concave or convex curve, and in a compres-
sible material, decidedly lacking in elasticity, this might easily lead
to the formation of fissures.

A very noteworthy point about these fissures, formed away from
and independent of, the river courses, is the manner in which they
usually run parallel to, and along either side of, any road or embank-
ment, [t is a well known mechanical principle that where there is
a sudden change in section, in other words a sudden change in
strength, the point of junction is a point of weakness, and
the strength at that point is less than it would have been if one
part had not been stronger than the other. For instance, in a
screw bolt, as ordinarily made, the portion on which the screw thread

! Brit. Ass. Rep., 1850, p. 53.
| ( 89 )

go OLDHAM: GREAT EARTHQUAKE OF 1897.

is cut is weaker than the shank; if the screw thread ends
abruptly the bolt will always give way at the junction of the two
parts, this being a point which is weaker than any other in the bolt,
and, paradoxical as it may seem, that bolt may actually be made
stronger by weakening the stronger part till the strength of the bolt
is equal at every point of its length.

This principle is of universal application, and from it we see
how the heaping up of an embankment on the surface of the
alluvium produces a line of weakness along its base on either side,
Besides this cause of weakness, we have the fact that the material .
for the embankment is usually derived from a row of borrow-pits
on either side of the bank. These two causes acting together give
rise to a special aptitude for fracture, and we find nearly every-
where throughout northern Bengal and lower Assam that the
roads and railway lines were bounded on either side by a set of
fissures running parallel to the road in all such places where
fractures have not, for other reasons, been formed across the road.

There was a very noticeable tendency for fissures, of the nature
now being considered, to range themselves parallel to each other,
and at first the direction was not unnaturally presumed to be at
right angles to the direction of propagation of the earthquake wave.
This assumption was, however, found to be untenable and no such
general rule applicable. Whether any connection might have held
good between the direction of the fissures and that of the visible
surface waves cannot be decided, as the observations are too scanty.
It is certain that these undulations did not necessarily travel in the
same direction as the earthquake wave,! and it is possible that where
the movement was sufficient to cause fissures to be formed, they
followed the direction of the crests and hollows of the surface
waves. Even if we suppose, what is probable enough, that the
direction of propagation of these waves is largely governed by local
peculiarities of the constitution of the surface layer of the alluvium,

1 See Chap. ll, pp. 20, 26, 36, 40.
( go )

FISSURES AND VENTS. QI

these same peculiarities would encourage the formation of fissures
in the same direction as the crest of the waves, or at right angles
to it, rather than in any other direction.

In his official report on the earthquake in Kuch Bihar, Mr.
D. R. Lyall, C. S. I., Superintendent of State, remarks that where the
soil is sandy the usual form of fissure was a simple opening of the
soil, but in the racecourse at Kuch Bihar, where the soil is more
clayey, there was generally a double fissure, the soil between which
sank from a foot to eighteen inches, and, as a rule, no sand issued.
Near Maimansingh Mr. Grimes noticed the same feature, where
fissures ran on either side of the road, and throughout the area
affected by fissuring, it is common to find two parallel fissures bound-
ing a strip of sunken land.

Where the fissures penetrate down toa bed of sand or slime,
part of which has been forced out through the fissures, it is
easy enough to account for this subsidence, but it may also
take place where no material has been forced out from below.
Suppose two parallel, or nearly parallel fissures starting from the
surface and gradually approaching each other underground so as to
cut off a wedge shaped piece of alluvium as shown in fig. 10, then,
as the wave reaches such a position that we have opposite semi-
phases of the wave on either side of this pair of fissures and the
movement of the wave particles away from the fissures, as indicated
by the lower pair of arrows in fig. 10, there will be an opening of the

> \ Jeca
بے‎ \ / —
V

Fig. 10. Diagram to illustrate the formation of fissures in alluvium.
fissures and the wedge will drop down. As the wave passes on the
action is reversed, the motion of wave particles in both semi-
phases will be inwards and the wedge will be compresssed, as indicated

(91)

92 OLDHAM: GREAT EARTHQUAKE OF 1897.

by the upper pair of arrows. The waves pass on, the first condi-
tion is resumed, the wedge drops a little further, to be com-
pressed once more, and the process is repeated and the subsidence of
the wedge continues till the alluvium has attained its maximum
possible compression or the earthquake motion ceases, whichever
occurs first.

Another form of fissure, which is much akin to that just treated,
in as much asit is due to compression of the alluvium and not to
bodily movement at a free surface, is the fissure which so constantly
runs along the junction of hill and plain and is accompanied by a
settlement of the alluvium.

Al along the foot of the Khasi and Garo hills, wherever the
alluvium of the plains extends up to foot of the hills, and is not
separated by the fan of a local stream, the alluvium has separated
from the hill, and a sudden drop of from 1 to 5 feet is found. This
scarped face of separation is very marked where the road from Tura
issues on to the plain of the Brahmaputra, and has been referred to
in official reports as a proof of subsidence. The vertical face of
separation is about 5 feet high and at first. sight has much the
appearance of a true fault, but on closer examination is found to
follow the windings and irregularities of the foot of the hills in a
manner which makes this explanation impossible.

The separation along the foot of the hills is not confined to the
alluvial plain of the Brahmaputra but is also noticeable in the case
of the small patches of peaty alluvial soil found in the valleys on
the plateau of the Khasi hills. These alluvial bottoms are carefully
terraced for rice cultivation, up to their very edge, and as the Khasias
had commenced to irrigate them when I marched through the hills
in March, I was able to see at a glance, by the varying depth of water
lying on the land, what changes of level had taken place.

[n many places I noticed that the original curve of the horizontal
alluvial surface into the hill slope was abruptly cut off by a per-

( 92 )

FISSURES AND VENTS 93

p endicular drop, as at bc, fig. 11. From Û outwards from the hill was

0

Fig. 11, Diagram to illustrate the displacement of alluvium alcng the foot of a hill.

a depressed area of from ro to 20 feet broad in which the water had
accumulated and completely flooded the fields ; immediately outside
this was a strip, d which had evidently been raised above its original
level, for the water could no longer be led onto it, and beyond this
came the alluvium, ‘preserving its original level and apparently
undisturbed, except where it has been cut by fissures.

These features clearly reveal what has taken place. The thrust
of the hill and plain against one another has caused the alluvium to
be thrust forward and raised into the low ridge or roll at d, while on
the return a space was left which could «not be filled up by the
alluvial soil without its surface being lowered. As in the case of the
depression of a wedge of alluvium referred to above, the action was
doubtless not completed by a single movement, but was the cumula-
tive result of a succession of waves of compression. |

On the borders of the alluvial plain of the Brahmaputra cul-
tivation does not extend right up to the edge of the hills, and we
lose the datum level afforded by the ricefields, but the same action
took place there also, and produced the same result. In no case was
the subsidence other than local and confined toa very narrow strip
of land immediately bordering the plane of separation; in some
places, as on the road from Tura, the amount of subsidence is in-
creased by the large quantities of sand extravasated, and the impres-
Sion of a general subsidence was favoured by the exceptionally high

( 93 )

94 OLDHAM. GREAT EARTHQUAKE OF 1897.

floods which rose right up to the foot of the hills, over land they
never covered before. As will be seen in Chapter IX these may be
accounted for without invoking any change of level, and, though it is
not impossible that there may have been some permanent change of
level it certainly did not take the form of a differential change
just at the foot of the hills, nor is it reasonable to invoke an ex-
planation of, which could not be applied to the precisely similar
appearances observed elsewhere,

In the foregoing passages the opening of fissures, where they
cannot be explained directly by a movement of the superficial
portion of the alluvium towards an excavation, whether natural or
artificial, has been attributed to a direct compression of the alluvium
consequent on the momentum imparted to it in each semiphase of the
wave, or to the effect of hard rock in contact with the softer alluvium.
This has been done as the explanation seemed the most natural and
simplest, but it must be pointed out that this same compressibility
of the soft alluvium may permit of a general movement of the
alluvium on a larger scale than that indicated; and several in-
stances of this have been reported. |

When the interrupted telegraphic communications were being
restored, the repairing. parties found that in many places the (ele:
graph poles, which had originally been set up in a straight line, were
displaced to the extent of ten or fifteen feet, and it was not always
possible to trace any connection between this displacement and
the neighbourhood of river or stream channels. "The most conspicu-
ous and best established instance which has been reported is on the
Assam-Bengal Railway. Mr, Grimes’ account of this will be found
in Appendix A and it is also referred to in the Inspection report of
Mr. J. E. Dallas, Offg. Senior Government Inspector of Railways,
Assam, in the following words :—

“The worst case of the lateral shifting of banks occurs at mile 1633. The
centre line in this case was shifted to a maximum of 6 feet و‎ inches; the length

(94)

a IINE T‏ فی یں پر

FISSURES AND VENTS. 95

affected is several hundred feet. I am not here describing the shifting of the
permanent way on the bank, or of the bank on the ground. On examining the
relative position of embankment, berm, borrow pits and trees, it is evident that
the lateral shifting is due to the displacement of the ground and not of the bank
only : there is no fissure at the foot of the bank, no wrenching of one part from
the other. At the points where the displacement begins and ends the bank was
much disturbed, practically wrecked, but there was no clean shearing anywhere.”

From Mr. Grimes’ account, it seems that this cannot be accounted
for by a movement of the alluvium towards a river course, for the
movement is parallel to the only river course in the neighbourhood ;
it is more- probably attributable to the local presence of some
yielding bed underground, along whose upper surface the superficial
layers were able to move, the movement being taken up at the
margins of the displaced tract by a gradual compression, extension,
or distortion of the alluvium in a manner that would not have been
recognisable but for the accident of the railway line crossing this
disturbed tract.

Another effect of this shifting of the alluvium, due to its power of
yielding to compression, was to be seen throughout large areas of
Northern Bengal, Lower Assam and Maimansingh. Here the rice-
fields, which are most carefully levelled to allow them to be flooded
to a shallow and uniform depth, were found, after the earthquake,
to be thrown into gentle undulations with a difference of level of
occasionally as much as two or three feet between crest and hollow,

and this apart from any changes of level directly due to the opening
of fissures or sand vents.?

The displacement of the piers of bridges might be attributed to
a similar cause. I do not refer at present to the displacement of the
abutments of the bridges, or the shortening of their span, which is
everywhere noticeable in the fissured tracts, but to the lateral

1In reply to an enquiry, the Chief Engineer of the line stated this to have been
2,400 feet.

2 The disturbance, which even comparatively deep-seated beds in the alluvium were
Subjected to, is evidenced by the fact, recorded in Mr. D. R. Lyall's official report to the
President of the State Council, Kuch Bihar, that eleven bends were found ina single length
of pipe in a Norton’s tube well when it was taken out after the earthquake.

( 95)

96 OLDHAM: GREAT EARTHQUAKE OF 1897.

displacement of the piers whose foundations are sunk in the beds of
the rivers and could not consequently move laterally in the manner
indicated in fig. 8. Instances of such shifting are mentioned in
Mr. Hayden’s report (Appendix A), and the effects in one instance
are represented in Plate VII, where it will be seen that the piers
have been shifted in an up and down stream direction, and also
along the length of the bridge in such manner as to deprive the ends
of the girders of their proper support. Another instance is shown in
Plate XXVI, fig. 2, where the narrowing of the space between the
abutments is made very conspicuous by the bending of the hand rails
and the projection of the girder over the abutments, beyond its

original bearings.

Though this movement of the piers might be attributed toa
general movement of the alluvium, such as that recorded on the
Assam-Bengal Railway, it is more reasonably attributable to the dis-
turbing effect of the forward thrust of the banks, which has disturbed
the equilibrium of the alluvium under the river bed and caused its dis-
placement, even to the depth of the foundations of the bridge piers.

The pushing forward of the abutments of bridges was universal
throughout the fissured tracts and is referred to in Mr. Hayden’s
report and that of Mr. La Touche, and the narrowing of the river
channels seems to have been still more conspicuous in places where
there were no massively built bridges to assist in supporting the
banks. In all cases it is due to the throwing off of the unsupported
river banks and accompanied by the formation of fissures. As it
was important to determine whether this narrowing of the river
channels was in any way connected witha general compression of
the-country, enquiries were specially instituted with this object, but
the replies received from the engineers in charge of the railway lines
were unanimous in saying that the narrowing of the river channels
had always been accompanied by breaks in the line on either side of

an equivalent amount.
( 96 )

e A‏ وھ را ہی

FISSURES AND VENTS. 97

A very striking result of the displacement of the alluvium, whether
by throwing off at the free surfaces of river channels or tanks
or otherwise, was the bending of rails. This has been reported and
illustrated before, especially in the case of the Charleston earthquake
of 1886, and the great Japanese earthquake of 1891, but never has it
been exhibited om so large @ scale, or over so large an area as in the
earthquake of 1897. The splendid series of photographs taken by
Messrs. Kapp & Co., of the effects of the earthquake on the Assam
Bengal and Eastern Bengal Railways, show many instances of this,
but none so striking as that on the Tezpore-Balipara tramway
reproduced on Plate VI from a photograph for which I am indebted
to Mr. Kemlo.

This bending of the rails is uncontestible proof of compression,
as it means that the two ends of the rail are closer together than
they were before the earthquake, and here again there was an
opportunity of determining whether there had been any general
compression of the country or not. If the compression of the rails
was only a local manifestation of a general compression it would be
unaccompanied by a corresponding expansion, whereas. if the coms
pression were only local and due to displacements of the alluvium:
by the action. of the earthquake wave there should be, for each case
of compression, a corresponding stretching of the line elsewhere ir
the neighbourhood.

With the object of ascertaining: whether this. was the case or not
special enquiries were instituted, of the managers of the Eastern
Bengal Railway and of the Tezpore-Balipara tramway, as to. whether
each case of compression was accompanied by a corresponding
extension of the line near by, or whether any general shortening was.
noticed. Unfortunately repairs had already been carried out and.
detailed measurements could no longer be made, but the replies
distinctly point to each case of compression being, accompanied by.
breaks in. the line indicating an.expansion which was, at any rate
approximately, equal in amount.

El | | (977

ee

98 OLDHAM: GREAT EARTHQUAKE OF 1897.

Mr. J. W. de Tivoli, agent and manager of the Tezpore-Balipara
tramway, writes :—

* No measurements were taken at the time, excepting of the biggest gap
between the ends of rails where the line was torn apart. This distance was 5 feet
6 inches between the end of one rail and the other rail which had been fished
to it. This break occurred 73 yards (measured by pacing) from the bend in the
rails nearest the camera in the photograph (Plate VI) showing the two bends,
There were eight other breaks between Sessa and Rangapara (43 miles), the
break being in all cases from 200 to 600 feet from the places where the line was
bent out. In every case of a gap there was a corresponding bend or bulge within
a few hundred feet as above stated. All the damage done to the line occurred

where the subsoil was sand, saturated with water. Where the subsoil was compact.

red clay there was no damage whatsoever.”

From the authorities of the Eastern Bengal State Railway the
information received was of the same general tenour. Mr. J. A.
Anderson, Engineer in Chief, wrote :—

* From two sketches I took on the ground, with measurements, I can state

that in one case two rails closed up 1! 8" in 48’, and in another 3 rails closed up `

1' 4" in a distance of 72' ; in each case, besides alarge amount of pulling out of
joints, there were sheared fishbolts and gaps of 5" to 8" at each side of the com-
pressed portion. | |

There were several lesser contortions in each case and I would guess the total
shortening of rails, including closing up of joints to be about 2' 6" in 240' this
shortening being equated by the pulling out of some 30 to 40 joints on each side
and the gaps above mentioned of 5" to 8".

In ordinary line the joints could close up a little less than 1" each, and could

KEE
e

open out a little more than 4
Mr. G. Moyle, Superintendent of Works, wrote :—

* Platelayers had to cut out such a large number of bent rails and close up so
many joints that l am almost certain they kept no record that I can supply,
with as near an approach to accuracy as possible, the general effect of the earth-
quake on the permanent way. i

Cranking of rails and tearing of fished joints was frequent at miles 132-133
(Nilphamari), miles 1575 to 165 (Haldibari) ; still more frequent between Rangpur
and Kaunia, miles 126-145 ; very frequent throughout the trans-Teesta 2' 6"
gauge lines, on account oflight rails and small gauge. There wereonly two
cases of 5olb rails cranking—the rails adjacent to Nilphamari up points and
just beyond Nilphamari down-signal—and no instance of tearing. 1 only saw
one instance of the line buckling in one place without tearing open at another
within a few hundred yards, and that was at Raninagar, where the main line was
shortened about 12",

( 98 )

FISSURES. AND VENTS. 99

In all other cases cranking at one place was always accompanied by tearing
at one or othef, or on both, sides of the cranked rails, sometimes about 80 yards
away, but usually much further, and in a few cases upwards of a quarter of a mile
away. Asarule, rail joints, when torn, only opened about 3” but in some cases
they opened out considerably more, and in a few instances as much as 10” or 11”.
I do not remember any case of the rails being torn asunder without cranking
occurring in the vicinity.”

From these statements it is obvious that the shortening of the
rails does not necessarily indicate any general compression, and that
it is rather to be attributed to local displacements of the alluvium,
whereby a compression in one place is compensated by an expansion
elsewhere, and generally at about 300 yards away.

So far I have considered only the effects of the horizontal com-
ponent of the wave motion, or at any rate, the effects which might
have been produced had there been no vertical component of the wave
motion. We now come to the consideration of those effects which can
only be explained by supposing that the wave motion had a vertical
component, in other words, that the movement of the wave particle
was to a certain extent up and down as well as from side to side.

First of these come the sand vents, and the first thing to be
noticed is that in some cases they were not formed till after the
earthquake shock had passed away. On this point the evidence of
Mr. D. R. Lyall's account is decisive in itself", and a similar statement,
to be found in the reports of Captain Gurdon and of the Magistrate
of Bogra, who records that the 64orkas or sand craters appeared after
the earthquake had ceased?.

This appearance of sand vents, after the actual movement of the
earthquake wave had passed away, is easily explicable, as the result
of the momentum imparted to the water bearing stratum, a momenture
which could not be immediately absorbed, and was in some cases
sufficient to force the sand and water through the overlying beds
weakened, as they doubtless were in places, by the earthquake which
had passed away.

1 See p. 26. | | ? See Appendix B
H2 (99)

AI AA

100 OLDHAM: GREAT EARTHQUAKE OF 1897.

Less obvious of explanation is the fact, which appears well
authenticated, that the issue of sand and water went on in some places
for several hours after the earthquake was over. In the case of the
earthquake of Vostizzia, on 26th December 1861, Dr. J. F. J. Schmidt
observed that some of the vents then formed continued active for
two hours after the shock had passed, but the activity was in this case,
due, to the issue of gas, smelling strongly of sulphuretted hydrogen,
the product of decaying organic matter. In the case of the present
earthquake, too, some of the reported cases of continued activity of
the sand vents were due to the deceptive appearance produced by
the rising of gas bubbles through the water which filled the depres-
sions in the craters, but there is still a sufficient residuum of credible
accounts which point to a real outflow of water from the vents for
some hours after the earthquake, in some cases said to be for 24
hours.

In Captain Gurdon’s account he distinctly says that, the ‘sand
geysers were subsequent to the great shock and continued for quite
half an hour.’ Inthe Assam Government’s official report on the
earthquake, itis said that at Gauhatithe bubbling lasted for 24
hours; other accounts from different parts of Bengal, which it is
needless to quote in detail", mention the discharge as continuing for
12 or 24 hours, periods which evidently do not represent accurate
periods of time, but may be interpreted as some or several hours.

Captain Gurdon’s more moderate statement is, however, sufficient
to show that there are cases in which the issue of sand and water

may continue for a longer period of time than is explicable by the `

momentum imparted to the sand and water, or the overlying imper-
vious layer, by the earthquake wave. This may most probably be
explained as follows. When a layer of clay’ or clayey silt begins to
form over loose sand, it at first exercises but little pressure, but as
the thickness increases by successive additions of silt, the pressure
on the underlying water logged sand increases. The specific
gravity of this compact silt is greater than that of the quicksand

( 100 ) 1 See pp. 16, 20, 25, and Appendix B.

p

FISSURES AND ۰ iot

below and, were there a free outlet, it would sink into or on to the
sandy layer till this had been compressed as far as the superin-
cumbent weight could compress it. In the normal condition of
affairs, however, the continuity of the surface layer allows of no
free outlet of sand or water. The earthquake not only opened
such outlets but, by the disturbance of the sand below, doubtless
loosened it and made it more mobile, and as the specific gravity of

the surface layers was greater than that of the water mixed with

sand which issued from the fissures, the blocks, into which the fotmer
was broken up, settled gradually down on to and into the underlying
quicksand, forcing up the superfluous water, which carried with it a
certain amount of sand.

It is probably to this settlement of the surface layefs that, in
many parts, there was a conspicuous absence of the deep crateriform
hollows due to the washing down of material by the indraft of the
water returning after the passage of the shock. The absence of the
scoring of the sides of the craters, so well shown in the photographs
taken after the Cachar earthquake of 1869 was due, doubtless, to
their obliteration by the heavy rains which fell immediately after the
shock, but the crateriform hollows, shown so conspicuously in Plates
X and Xl, are in many places wanting, and all that could be seen
were low rings of sand with a shallow hollow in the centre.

The force with which the sand and water issued from the vents
was evidently considerable. In many accounts it is vaguely stated
that the fountain rose to as much as 10 or 15 cubits, but these
may be rejected, unless the phrase is meant to apply merely to stray
splashes, which may easily have risen to that height. In the Charles-
ton earthquake of 1886, it was found that the branches and leaves of
trees standing over some of the vents bore splashes of mud up to
à height of 13 feet above the ground,! and, considering the much
greater scale of this earthquake, it is easy to suppose that stray
splashes may. have risen to even greater heights. The main body

. U. S, Geol. Surv., gth Annual Report, p. 289 (1989*.

( 101 )

102 OLDHAM: GREAT EARTHQUAKE OF 1897.

of the water and sand, however, did not rise to anything like this
height, though there is ample evidence that in many cases it rose in
a solid column to a height of 2 to 4 feet or more.

Mr. R. R. Morgan, Superintendent of Pilots on the Brahmaputra
river informed me that he was, at the time of the earthquake, proceed-
ing up the river and was an hour’srun from Dhubri. The first sensa-
tion felt made him think the steamer had run aground, and on going
on deck he saw two spouts rising one on the north, and the other on
the south, whose height he estimated at 12 feet.

Other credible accounts do not place the height at so high
a figure, though this need not impugn the statement of Mr. Morgan,
as it is probable that locally and for a short time the water was
forced to a greater height than the average.

A few of these accounts may be noticed. At Nalbari, in the
northern part of the Kamrup district, Captain Gordon observed the
water rising to a height of from 3 feet to 4 feet; Rai Bahadur Madhub
Chandra Bardolai, Subdivisional Officer of Barpeta, reports that the
sand and water rose to heights varying from 2 to 12 ft.; in a report
by Devendra Bijoy Sur, Munsif of Isvarganj in Maimansingh, he
states that he saw several vents being formed and estimated the
height to which the water rose at 10 feet ; at Rangpur the vents were
seen in action by Mr. J. Elison, who observed them to rise in a solid
column of sand and water to a height of about 1 feet 6 inches
to 2 feet. In chapter II, there will be found accounts of eye-
witnesses who estimate that the sand and water rose to heights
varying between 2 and 4 feet.!

From these accounts we see that, omitting stray splashes or
exceptional spurts, the bulk of the sand and water was forced not
only to the surface but in a solid stream to heights of 2 to 4 feet above
it. The force with which,the water issued from these vents is also
exemplified by the solid substances brought up with the sand. From
almost every district where sand vents were formed to any great

1 pp. 15, 20, 25, 26; see also Appendix B.
( 102 )

FISSURES AND VENTS, : 103

extent, it is reported that trunks of trees were thrown out, and from.
places near the foot of the Garo Hills the presence of lumps of coal
and fossil resin is recorded. The specific gravities of these sub-
stances is; however, low, and they might be raised, even in pieces of
considerable size, by no more rapid a current than was necessary to
raise the sand.

This explanation will not, however, apply to two recorded cases
of heavy substancés being ejected. In the north of the Tezpur
district, near the Bhoreli river, waterworn pebbles of hard rock were
thrown out with the sand, and at Haldibari, in Northern Bengal,

lumps of “lava”

weighing as much as half a pound were thrown out
with the sand. On examination this proved to be what is known
in India as 7Aama, the overburnt bricks from the lining of the fire-
chambers in brick clumps; it had undergone a marked decomposition
and the fragments had acquired a whitish coating; they were
evidently the remains of an old brick kiln, which had long been buried
and were washed up with the sand and water as it issued from the
ground.

In the inspection notes of Lieutenant Colonei H. St. P. Maxwell,
C.S.I., Officiating Commissioner of the Assam Valley districts!, two
_ instances are mentioned which, if accepted as they standy would lead
to a very exaggerated idea of the force with which the sand and water
issued, and for that reason must be noticed here. The first of these
is at Nowgong, where it is said that the ejection of the sand had
such force that covers of wells; imbedded in mortar, were hurled aside,
The second is at Goalpara, where it is said that a well was altogether
filled with sand and a portion of the wooden cover was hurled 30
feet distant where it lies half buried.

The first mentioned instance I was not able to examine person-
ally, but the statement that the force of the issuing sand and water
was sufficient to detach the wooden well covers may be accepted
. as correct; the other statement, regarding the well at Goalpara,

! Printed in the appendix to the official report of the Government of Assam,

( 103 )

104 OLDHAM: GREAT EARTHQUAKE OF 1897.

is also correct if taken literally, but not in the interpretation
which would naturally be put on it that the well cover was hurled
through the air to its present position; while the true facts are, in a
way, even more remarkable than Colonel Maxwell’s description
would indicate. ;

The well in question was a brick well surrounded by four brick
pillars 5 feet high by 1 feet 3 inches square, which supported a wooden
framework carrying pulleys for convenience in drawing water. Of
these three remain standing, though cracked, while the fourth is
missing. The well stands just on the line of one of the fissures
which opened in the Goalpara bazar, and from which large quantities
of sand were poured forth, This issue of sand was naturally espe-
cially abundant at the well, and the flood of sand and water picked
up the broken pillar and floated it 65 feet to N. 50° E., from its
original position to the spot where it now lies. The wooden well-
cover I did not see, as it had probably been removed before my
visit.

It will be seen from this that we have not to do with an actual
projection of the masonry, and probably not of the wooden well-
cover either, but still the force of the rush of sand and water must
have been great to enable it to pick up a mass of brick-work of about
6 cubic feet in volume and carry it 65 feet from its original position.

Closely connected with the formation of sand vents was the
filling up of river channels, tanks, wells, and all other depressions or
excavations, over a large area. In part this was due to an actual
outpouring of sand, but to a far greater extent to the forcing up
of the bottoms of the excavations, That this is the true explanation
especially in the case of light, bamboo structures, is clearly shown
by many bridges crossing small streams or canals, these have in
innumerable instances been forced up in the centre (see Plate XXVI)
and the roadway, which was once horizontal or neaily so, rose after
the earthquake in a more or less steep slope from either side, This

( 104 )

FISSURES AND VENTS. 105

forcing up of the central piers of the bridges shows that the filling
up of the channels was due to an actual raising of their beds, and
not to the outpouring of sand through vents and fissures.

The mechanism of this effect is easy to understand. If, instead
of the continuous stratum represented in figs. 8 and 9, we suppose it
partially cut through, as in fig. 12, by the tank T, the well w or

WY کر‎ : LZ

Fig. 12. Diagram to illustrate the filliag up of stream channals and excavations.

the stream bed R, it is obvious that these introduce areas of special
weakness and that when pressure is brought to bear on the bed of
loose sand it is at these spots that the overlying layers of alluvium will
most readily yield. Owing, however, to the area over which yielding
takes place, it will often happen that pressure is completely
relieved without any actual outpouring at the surface, and as the
same amount of relief can be given by a lesser actual rise where the
area is large than where it is small, the raising of the floors of
depressions will, ceteris paribus, be inversely proportional to the
area, This is in accordance with experience; the greatest raising
is usually observed in wells, and these have often been affected when
there has been no noticeable change in tanks or watercourses.
Next to wells narrow watercourses have been most affected, while
in large tanks and natural pools the raising of the bottom has been
less than in water channels of equal depth but smaller width in the
neighbourhood.

This filling up of river channels took place over a large area, but
probably nowhere so conspicuously as in the tract of lowland which
lies between the foot of the Garo Hills and the Brahmaputra. This
.tract is intersected by numerous channels, which carry a limited
drainage in the dry weather, but, when the Brahmaputra is in flood

( 105 )

ِ
|
|
i

t o6 OLDHMAM: GREAT EARTHQUAKE OF 1897.

help to carry off the surplus waters that would otherwise submerge
the country they drain. Before the earthquake these channels were
from 15 to 20 feet deep, and in the dry weather the country was inter-
sected by steep-sided depressions of this depth; at the bottom of which
flowed a shallow stream. During the earthquake the bottoms of all
these channels were forced up till level with the banks on either
side, and during the ensuing dry weather the drainage of this tract,
instead of flowing in deeply sunken channels flowed nearly level with
the general surface of the land in shallow sandy channels.

This blocking of the drainage channels prevented them from
serving their natural purpose when the rivers rose in flood, and large
volumes of water which would otherwise have escaped were forced to
spread over the surface of the land, thus giving rise to the exceptional
floods of 1897. This matter, being dealt with elsewhere}, will
not be further referred to, but a direct result of the upheaval of
the river channels may be noticed. As, this upheaval was not due to
an eruption of material from a great depth, but to a redistribution
of comparatively superficial material, the rise of the river beds had to
be compensated by acorresponding subsidence elsewhere, in general
immediately contiguous to the river banks. It is obvious, that in a
material which offers so great internal frictional resistance to deforma-
tion as the alluvium, the movement by which the deformation took
place would ordinarily be transmitted to the shortest distance com-
patible with the effect to be produced; hence for this reason alone

the subsidence of the banks would probably be greater than that of

the alluvium further back.

Added to this there are two facts to be considered: firstly that,
as the alluvium near the river banks was most broken up by fissures,
the separate blocks would subside more readily than a large continu-
ous sheet, and secondly that, as is the case with all deltaic rivers, the
river banks were higher than the land behind them, and the pressure
on the subjacent yielding layers greater than elsewhere,

1 Chap. IX.
( 106 )

FISSURES AND VENTS. 107

All these causes combined would cause the land immediately on
either side of the river channels to subside more than elsewhere, and
so the relative greater elevation of the river banks would disappear.
In fact, the general effect of the earthquake throughout the districts
where vents and fissures were formed, may be described as tending
to reduce inequalities of level by causing a depression of the high
ground and an elevation of the low,

A very remarkable result of the earthquake was the sudden rise
of the rivers which accompanied it over a large area, I do not here
refer to the extensive floods which were experienced later on and
were indirectly the result of the earthquake, but to a sudden
and immediate rise, varying from two to ten feet, which was observed
immediately after the earthquake and passed off in the course of the
next day or two.

At Gauhati the river gauge, which is fixed ona hill of gneiss,
showed a height of 67'41 ft. above sea level at 7 A.M. on the morning
of the 12th June; at 6 P.M., about three quarters of an hour after the
earthquake, the water stood at 175'00, showing a rise of 7'59 ft; the
next morning it had sunk to 173'11, at 7 A.M. on 14th to 169'96 and at
7 A.M. on 45th to 167'91, having thus recovered its normal level in
two and a half days.

At Goalpara the Brahmaputra river is said to have invaded the
station in the form of a wave 10 feet high and remained at 8 feet
above its previous level after the earthquake had passed ; at Kaunia,
the, Dharla river is reported to have risen 3 feet immediately after
the earthquake. At Maimansingh the Brahmaputrariver rose 4 feet
and at Jamalpur 5 to 6 feet, In the Bogra district it is reported that
the rivers all rose from 1 to 4 feet and that they all fell again within
the course of the next three or four days; and so on. Many more
reports could be quoted, but sufficient has been done to show that
_ the phenomenon was a general, not an isolated, one,

In part this rise of the rivers may be attributed to the large

( 107 )

108 OLDHAM : GREAT EARTHQUAKE OF 1897.

volumes of water poured out from the sand vents. Surgeon Major
Dobson's account! shows clearly how this might affect the amount of
water in the drainage channels, and other accounts also describe how
streams that were nearly dry suddenly filled with the water which
issued from the ground. This càuse would, in many instances, account
for the sudden rise in the smaller drainage channels, but in the larger
streams the rise was too sudden atid followed too promptly on the
earthquake to be explained in this manner. Here we must look to
the raising of the stream beds to account for it, and to the gradual
scouring out of the barriers so formed for the subsequent fall in level.

The raising of the river beds was probably continuous to a
greater or less extent all along the channels, and the water was
lifted bodily to a certain extent by the earthquake, but the major
part of the rise appears to have been due to a different cause. The
forcing up of the river beds was by no means uniform, and in some
places was more extensive than in others; in this way barriers were
formed across the stream, and on the up stream side the water was
ponded up to the height of the maximum rise of the next barrier
down stream. Thus a greater rise of water level occurred than would
have been the case if the raising of the river beds had been uniform,
while the barriers, being composed of loose sand, were more easily
scoured away and the material of which they were composed dis.
tributed along the stream bed in such manner as to leave the water
lower than the level to which it had risen immediately after the
earthquake, though a little higher than it was before.

It was only where the blocking of the river channels took this form,
and where there was a strong current, with its consequent power of
scouring, that the channels reopened. Where the current was slack
the beds were not scoured out and remained choked when the floods
of August and September subsided ; and where the channel was con-
tinuously and completely choked, the streams have in many cases
deserted their old channels and formed new ones.

( 108 ) 1 See page 15.

FISSURES AND VENTS. 109

In the foregoing passages the bearing of the phenomena, which
were observed on the formation of earth fissures and sand vents has
been discussed; we must now consider their extent and distribution.
Before entering into this, however, it may be well to notice one
qualification which must be borne in mind whenever it is said that
they were universal or generally distributed over any specified area,
namely that they can only occur where the local conditions are
favourable.

The whole of the hill area must, of course be excluded, but even
in the alluvium there are large tracts of an older alluvium, charac-
terised by a much greater compactness than the newer. This
older alluvium is usually of a red colour and forms elevated tracts,
rising from the newer, of which the Madhupur jungle, lying between
Dacca and Maimansingh, is the best known, though not the only one.
To the north of the Ganges and in the Assam Valley there are severa]
tracts of similar alluvium, which have everywhere escaped fissuring
or the formation of sand vents, owing to their greater compactness
and the absence of beds of loose waterlogged sand. Even in the
newer alluvium such beds of sand are not everywhere near enough
to the surface to allow of the production of the effects described in
this ‘chapter, but with these exceptions it may be said that fissures
and sand vents were universal throughout the Goalpara and Kamrup
districts, the western part of Darrang and the greater part of
Nowgong, Sylhet, and the northern parts of Cachar, In Bengal
they were equally prevalent throughout Rangpur, Dinajpur, Raj-
shahi, Maldah, Purniah, Pabna, Bogra, Maimansingh, and the greater
part of Dacca.

Outside the area over which they were widespread, there is a
larger area where they were less frequent. Fissures and sand vents
are said to have been formed in the Murang district of the Nepal
Tarai, north of Purniah. In Bhagalpur a few were found in the
Supul sub-division, and near Colgong, on the banks of the Kosi and
Ganges, respectively. A fissure 50 feet long by 1 foot broad is said to

( 109 )

110 OLDHAM: GREAT EARTHQUAKE OF 1897.

have opened in the Bihar cutchery compound. In the Santal
Parganas they were formed near Deogarh and Godda. Stray
fissures are reported to have been formed in the Bardwan, Bankura,
Birbhum, and Midnapur districts, but none in Hugli or Howrah.

Fissures and vents are also reported from the Murshidabad, Nadiya,
Jessor districts, none were formed in Faridpur or Bakarganj, or in
the Chittagong division, with the exception of the extreme northern
part of Tipperah. |

In Assam fissures are reported from as far east as the Lakhim-
pur district, but only to the west of the Subansiri river. In the
Sibsagar district a few occurred in the western part of the Sibsagar
sub-division, but none to the east of Sibsagar. They were formed in
the alluvium of the Dhaneswari (Dhansiri) valley, but none in Mani-
pur. In Cachar they are confined to the north-western part of the
district, and they were formed all over Sylhet.

From this it will be seen that the area over which fissures were
fairly frequent, wherever conditions are such as render their forma-
tion practicable, corresponds very closely with that included in the
fifth isoseist on map No. I, while the fourth isoseist includes the
most distant places at which fissures could be formed even where
circumstances were most favourable. In other words, the area over
which fissures were formed abundantly, or would have been if local
conditions were not adverse, reaches about 400 miles from east to
west and about 350 miles from north to south, while the extreme
limits from which they are reported to have been formed was nearly
600 miles in an E. N. E.—W. S. W. direction, between Sibsagar and
Bihar ; and 300 miles from north to south, between the Nepal Tara
and Midnapur.!

The greater length in an east-west direction is, however, de-
ceptive, for it is partly, and may be wholly, due to the great range of

1 Fissures were reported to have been formed in Gwalior and in the Kistna district. On
closer enquiry being made, it was found that no connection could be traced between them
and the earthquake beyond the fact that they opened somewhere about 12th June. They were
probably due to drying or slippage of the soil and would have attracted no notice had it
not been for the earthquake.

( 110 )

FISSURES AND VENTS. 111

the Himalayas, which not only limits the area over which the formation
of fissures and vents could be possible, but places an absolute barrier
against the acquisition of any information as to the effect of the
earthquake in that direction.

` CHAPTER VII.—THE LANDSLIPS.

Landslips were caused by the earthquake on an enormous scale,

and deserve a special notice, both as to their origin and distribution,

The cause of the landslips is not difficult of explanation. In
fig. 13 let the shaded portion represent the solid rock of a hill, drawn

Q» l

Fig, 13. Diagram to illustrate the formation of landslips.
in section, covered by the soilcap O O which is kept in place on the
underlying rock by the friction between them. When the rock of
this hill is set into elastic vibration by the earthquake wave, the
superficial portion will, at one period or other of the shock, be set in
movement outwards, and this movement will be communicated to the
soilcap. In the next semiphase of the wave, the movement of the
surface of the rock will be inwards, but the inertia of the overlying
soilcap will prevent this following at once, and the effect will be a
more or less complete reduction of the pressure of the soilcap on the
rock, This reduction of pressure means a reduction of the friction,
which alone prevents the soilcap from sliding bodily down the hill,
and so alandslip is formed, where the reduction of resistance, and the

slope of the hill, are sufficient to allow of it. In cases where the

( 11x )

B12 OLDHAM: GREAT EARTHQUAKE OF 1897.

adhesion of the subsoil to the underlying rock is great, where its
thickness is small, or where the violence of the shock is not great
enough, the slippage of the surface layer does not amount to a land-
slip, and in these cases the hillsides are found to be scarred with
fissures. At the other extreme, the momentum imparted to the
surface layer may be so great that an actual outward pull is set up,
sufficient to overcome the resistance, both of gravity and the adhe-
sion of the subsoil to the rock, and then we have not merely a land-
slip, but the whole face of the hill shot bodily off. The conditions
that allow of this appear to be rare, but they certainly did occur in
some of the high scarps of the Khasi and Garo Hills.

So far I have spoken only of the landslips produced by the dis-
placement of the more or less weathered surface layer ; they seldom
extend deep into the hill, and resemble in all their characters ordi-
nary landslips of the same scale. Besides these, the high sandstone
scarps of the southern edge of the Khasi and Garo Hills exhibit land-
slips due to the throwing off of a greater or less width of solid sand-
stone, in a manner precisely analogous to that described in Chap-
ter VI but on a larger scale.

In the case of both these forms of landslips, the part of the hill
which is left standing is always scarred with deep fissures, extending
more or less parallel to the free face of the fall, and due to a partial
detachment of the material between them and the edge of the actual
slip. These fissures are very conspicuous on the Bálpakrám
plateau west of the Mahádeo valley, as shown in Plate XXVII, fig. r,
but they are in all cases a secondary effect of the transmitted earthe .
quake wave, just as the sand vents and earth fissures treated in the
last chapter.

Having noticed the mode of formation of the landslips, we have
now to consider the limitations on their formation. Foremost among
these must be placed the violence of the earthquake, for it is only
under very exceptional circumstances that a wave of low velocity, or
acceleration, of wave particle could give rise to landslips, Given a

Wie,

THE ۸0 117۰ 113

sufficiently violent earthquake the other factor that comes into play is
the natural tendency of the hill to slip. This obviously varies with
the slope, a gentle slope being much less liable to slip than a steep
one and the nearer the slope reaches that eritical angle, at which the
soilcap would slip away of its own accord, the less the impulse required
to set it in motion.

There is, however, another factor which appears to be of almost
equal importance with the angle of the slope, and that is its height
from base to crest. While travelling in the Assam hills during the
cold weather of 1897-98, I had many opportunities of observing how
steep slopes and scarps of a low height, had remained uninjured,
while gentle slopes, when forming part of a hill rising to several
hundred feet had been scarred with landslips.

This is well seen on the road from Shillong to Gauhati ; for the
first 14 or 15 miles, where the hills are high and cut by deep valleys,
landslips are common, but for the rest of the way, where the height of
the hills above the valleys is much smaller, landslips are almost un-
known though the earthquake was on the average equally severe
over the whole of this tract,

The explanation of this connection between the height of the
hills and the degree to which they have been scarred by land-
slips is doubtless the greater swing which was imparted to the
higher hills, In part, this is due tothe fact that an equal angular
motion would result ina greater linear movement at the top ofa
high hill than of a low one, but mainly it is due to the greater
elastic play of the high hill, especially when steep sided; just as the
end of a long switch jerked to and fro descrtbes a larger are than
that of a short one.

There is yet another factorin the production of landslips, and
that is the mineral constitution of the hill. When this is composed
of crystalline rock, the surface layer of disintegrated and weathered
material is either thin, or it passes down into the unweathered
rock in a gradual manner. In the former case, the thin skin may

(7و2ع:) 1

114 OLDHAM: GREAT EARTHQUAKE OF 1897.

not acquire sufficient momentum to cause it to be detached from the
underlying rock; in the latter, the gradual increase in the cohesion
of the surface layer adhesion to the underlying rock is, more or, and
in its less, proportionate to the increase of strain applied, and the
surface layer will be much less liable to come away than if there was
a more or less well defined plane of weakness.

In the case of sedimentary rocks, the boundary between the
weathered soilcap and the underlying rock is generally more abrupt
and the surface layer readily separates from the rock below. Added
to this the sandstones, which form so large a proportion of the hills
along the southern face of the range, have a much lower cohesive
strength than the crystalline and metamorphic rocks of the central `
and northern portions of the range and, when they form high scarps,
portions of the solid rock itself may be thrown off,

I have thought it necessary to devote some space toa general
consideration of the causes which are concerned in the formation
of earthquake-landslips, as very erroneous conclusions have been
drawn from their distribution, and are generally held in Assam, as to
the position of the focus of this earthquake. We see that besides
the energy of the earthquake wave, the production of landslips is
controlled by the petrological composition of the hill and by another
factor, into which the angle of slope and the height of the hill enter,
in other words, by its size and form. It is not possible to express
mathematically the relative importance of these different factors, the
nature of their influence can only be pointed out in general terms,
but itis evident that any conclusion drawn by attributing the whole
of the effects to only one of the operating causes must necessarily
lead to error,

After these preliminary remarks we may proceed to a considera-
tion of the distribution of the landslips and of their effects. In doing
this, it is necessary to remember that the observations and informa-
tion, on which the following passages are based, was obtained: some

( 114 )

THE LANDSLIPS, 115

time after the great earthquake, and that a part of the effects noticed
was due to the action of the heavy rains of the three months following
the earthquake, and part to the aftershocks of the great earthquake.
Many of these were themselves violent enough to cause landslips,
the more so as many hillsides, which had not come down in the
great shock, had been badly shattered and weakened, and were
more easily broken down than would otherwise have been the case.!

The heavy rains which followed the earthquake had probably
even a greater effect than the aftershocks in bringing down hill-
sides, which had been fissured and weakened by the earthquake, and
the total results observed and recorded after the close of the rainy
season included those of both these causes, nor was it found possi-
ble to separate the direct effects of the earthquake from its indirect
effects, or those of the aftershocks. It may, however, be safely said
that had the earthquake occurred three months later, at the end
instead of the beginning of the rainy season, and at a time when all
the hills were waterlogged, the landslips then developed would
certainly not have been less extensive than those which were actually
observed during the cold weather of 1897-98.

.. It seems certain too, so far as I could gather from careful enquiry,
that by far the greater number of the landslips which actually
occurred were the direct result of the great earthquake, and that the
subsequent additions to the size and number were only a fraction of
what could be seen immediately after the first shock,

Landslips were developed more conspicuously along the southern
edge of the Garo and Khasi Hills, and especially in the neighbour-
hood of the Paniathit river and the hills immediately to the west
of it. Viewed from the deck of a steamer sailing up to Sylhet,
the southern face of these hills presented a striking scene. The
high sandstone hills facing the plains of western Sylhet, usually
forest-clad from crest to foot, were stripped bare, and the white sand

1 On the 23rd of August, the Deputy Commissioner of the Garo Hills records in his tour

. diary that there were several severe shocks of earthquake, some of which sent rocks rolling
down the hillside. ;

E24 | C Uia)

116 OLDHAM: GREAT EARTHQUAKE OF 1897.

stone shone clear in the sun, in an apparently unbroken stretch of
about 20 miles in length from east to west. Beyond this, the land-
slips, though still large and conspicuous grew gradually fewer, and
it was not difficult to suppose that we had here the centre of the dis-
turbance.

The true explanation, however, lies in the physical features of the
hills, rather than in any greater violence of the earthquake. It
is in about g1? E. Long, that the hills rise highest and most abruptly
from the plains, and, being composed largely of soft nummulitic and
supra-nummulitic rocks, are more easily broken into landslips. To
the west, a strip of low hills separates the high hills from the plains,
and to the east the height of the hills decreases in a marked manner
beyond Cherrapunji. Eastwards of this I have not been myself,
but according to accounts received the landslips, though still fairly
numerous, are much less so than those near Cherrapunji.

At Cherrapunji the deep valleys are scored by landslips to a
striking degree, so much so, that when looked at from a distance
there appears to be more landslip than untouched hillside. Ft is
difficult to detect any relation between the size and frequency of the
landslips and the direction of slope, but though they seemed
equally abundant on the east and west sides of the valley, they
did seem to be somewhat more abundant on the slopes facing south-
ward than on those facing north.

Further north, towards the centre of the hills, no such relation
could be detected; the landslips were practically confined to the
sides of the deeply cut valleys, whose general direction is about north
north-east, south south-west. Inthe granite area round Laitlynkot
(Lailang Kot) granite tors had been thrown down in great numbers
and lay at the foot of the slopes. All this region has been extensively `
washed for iron ore in the old days, and most of the tors, which were
overthrown, had been more or less undermined in the process,

Throughout the high plateau of the Khasi Hills, to the westward
of Shillong, landslips are scarce, as the hills have mostly gentle

(. 110/79

THE LANDSLIPS. 117

slopes and do not rise high above the valleys. When, however,
there is a high range, like that of the Mauterichán or Laubersát,
which rises in a steep slope of 2,000 to 3,000 feet, or where the high
plateau is cut by a deep gorge, landslips are not uncommon; and on
the steep sandstone scarps of the southern face they are large and
conspicuous, as has already been remarked.

Nowhere that I know of are they so strikingly developed as in
the small valley of the Mahádeo, to the south of the Bálpakrám and
Pundengru Hills. The valley forms an amphitheatre of about four
miles long from east to west, and a mile and a half across, On the

north are the hills of Bálpakrám and Pundengru, united by a

narrow ridge of sandstone, scarped on each side ; to the west is
another narrow ridge of sandstone, uniting the Bálpakrám hill to the
plateau which separates the upper Mahádeo valley from the plains ;
at the eastern end, one of the spurs of Pundengru runs southwards
towards the outlet of the valley to the plains of Sylhet,

Here everything combined to favour the formation of landslips.
The hills were composed of soft sandstone, they were steep sided,
high, and narrow from side to side, and consequently were doubtless
thrown into actual oscillation as a whole ; while the range of motion
of the wave particle was not less than eight inches near the edge of
the precipices, The result of these local conditions, added to the in-
dubitably high energy of the earthquake wave, has been to produce
an indescribable scene of desolation. Everywhere the hillsides facing
the valley have been stripped bare from crest to base, and the
seams of coal and partings of shale could be seen running in and
out of the irregularities of the cliffs with a sharpness and distinctness

-which recalled the pictures of the cañons of Colorado. At the

bottom of the valley was a piled up heap of debris and broken trees,
while the old stream course had been obliterated and the stream
could be seen flowing over a sandy bed, which must have been

_ raised many feet above the level of the old watercourse.

At one place on the watershed, and on most of the narrow spurs

( 117 )

118 OLDHAM: GREAT EARTHQUAKE OF 1897.

running into he valley, the landslips on either side of the ridge had
united to form a sharp edged crest to the ridge ; and at one spot
on the ridge connecting the Bálpakrám and Pundengru Hills, where
a narrow Strip of the old forest clad summit had been left, the vio-
lence of the oscillation it had undergone was such that the forest
had been levelled to the ground.

This was not the only instance of this sort, but wherever
throughout the sandstone plateau, there was a high spur with steep
sides and a comparatively level forest clad top of a couple of
hundred feet or less in width, I noticed that the trees had almost
invariably been uprooted. They were not snapped across, like
those described in Chapter IX, the conditions which allowed of this
form of destruction being altogether local, but had been bodily
up-rooted by the swaying of the trees from side to side.

In the country on either side of the boundary of the Garo and
Khasi Hills Districts landslips are everywhere conspicuous in the
sandstone country and are decidedly larger and more abundant on
the southern than on the northern faces of the hills, but no conclusion
can be drawn from this, as to the direction of travel of the earthquake
wave, since the geological and orographical conditions are such as
to necessitate the highest and steepest slopes facing southwards.

Westwards of the Sameswari (Sumesari) river, the Garo Hills,
instead of rising in a high scarp from the plains, descend gradually
by a series of low hills, and the conditions are less favourable for
the formation of landslips, We find them, accordingly, less con-
spicuously developed, but wherever there is a high slope, such as the
southern face of the Tura or of the Arbela ridges, landslips are
numerous, while throughout the Garo Hills small ones are found on
the steeper slopes. —

The eastern limit of landslips appears to be in the North Cachar
Hills, as the reports specially called for state that none were formed
in Manipur or in the Naga Hills Districts. In the latter district
cracks in the hillsides, or incipient landslips, were formed near

( 118 )

THE LANDSLIPS. 119

Kohima, and it is probable that the same might have been observed
in the hills of the western part of Manipur, from which it is im-
possible to obtain any information. No landslips were observed
in the Mikir Hills east of Nowgong, or in the hills south of the Sylhet
valley.

The great range of the Himalayas with its deeply cut, steep sided
valleys is always a region of landslips. Every year, in a large number
of places, the strain becomes too great and a larger or smaller mass
of soilcap and partially weathered rock is detached, and crashes into
the valley below. In such a region, where there are large areas of
hillside in a critical condition, and the restraint on the movement of
the soilcap almost in equilibrium with the downward pull of gravity,
the effect of a severe earthquake may be expected to be great ; and
such was in fact the case. All along the north side of the Brahma.
putra Valley, the range of the Himalayas is scarred by landslips, even
to beyond Tezpur, In all years some are seen, but after the rains of
1897 they were in most unusual abundance, and many of them evie
dently of great size. As seen from the deck of one of the Brahma-
putra steamers, they are abundant even to the north of Tezpur, but
the effect of the earthquake does not appear to be very marked
much eastwards of this. Westwards from the meridian of Tezpur,
they get more numerous, and, so far as can be judged from the
reports, attain their maximum to the north of the Goalpara District,
in a part of the range not visible from the Brahmaputra. In Bhutan,
landslips are reported to have been numerous and extended far back
into the hills. They were observed in north-eastern Sikkim, and on
the south face of the hills were formed even as far west as Darjiling,
but apparently not in Nepal.

It will be seen from this that the east and west range of the land-
slips was more extensive in the Himalayas than in the hills south of
the Brahmaputra Valley, and as there is no reason to suppose that
the earthquake wave would be more readily transmitted along the
Himalayas than through the crystalline roc ks of the Shillong plateau

| | ( 119)

120 OLDHAM: GREAT EARTHQUAKE OF 1897.

this greater extent must be ascribed to the greater natural liability
of the Himalayan hillsides to slip.

One more point may be noticed, that the region where they are
most numerous lies a little west of the 91° meridian, or nearly north
or north-north-west, of the region where they are most developed
in the Garo and Khasi Hills.

Before finally leaving this subject, it will be necessary to notice
some secondary effects of the landslips in modifying the forms of
the watercourses. In the sandstone country at the southern edge of
the Garo and Khasi Hills, the dislodgement of large bodies of wea-
thered rocks, and to no less an extent the consequent exposure of
hillsides previously protected by forest, caused enormous volumes
of sand to be cast into the streams. One effect of this was the form-
ation of great sand fans along the southern foot of the hills, where
they wrought great havoc in the fields, and more especially in the
orange and areca-nut groves, of Shella and other villages near the
southern edge of the hills. Every stream of any size has in this way
devastated many square miles of country, and the mischief is not yet
complete ; for not only are there still large quantities of sand not yet
removed from the landslips, but the bare faces of the hills will be the
source of fresh supplies of sand as they are scoured by the rain,
until such time as vegetation once more resumes its sway.

Within the hills the effect has been no less striking. The burden
of sand cast on to the streams was far greater than they could carry
along their old gradients, and everywhere the beds have been raised,
changing the whole character of the river channel in the process,
Ordinarily, the beds of these rivers, which are raging torrents
when in flood, consist of a succession of deep pools separated by
rocky rapids. After the rains of 1897, it was found that the pools
had been filled up, and the rapids obliterated by a great deposit of
sand, over which the rivers flowed in a broad and shallow stream,
In this manner the well known fishing pools of the Sameswariand the

( 120 )

THE LANDSLIPS. 121

Paniathit have been obliterated, and many smaller streams converted
from deep, clear, rocky mountain streams to a shallow spread of
muddy water, flowing over a sandy bed.

That this change is entirely an indirect effect of the earthquake,
and a direct effect of the landslips it gave rise to, is shown by
the fact that it is only within the sandstone area, where landslips
are numerous, that the beds of the streams are raised in this manner.
When we get further back into the hills, where landslips are few
and where the crystalline rocks exposed by them are not so weathered
or so easily washed down as the tertiary sandstones, and where, in
consequence, the additional burden cast on the streams was com-
paratively small, we at once enter on an area in which the stream
beds retain their original character unchanged.

Cases where streams were ponded up for more than a day or
two were rare. One instance I saw at a village called Chochalja
about 8 miles to the east of Damra, on the northern foot of the Garo
Hills. Here a landslip had come down and shot out about 400 yards
into the open valley, crossing the drainage channel and giving rise to
a shallow pond, which had been completely filled up with sand by the
end of January 1898. A much more remarkable instance was that
of the landslips near Sinya, to the east of Rambrai, which dammed
up the drainage of a large area for nearly three months and gave rise
to a destructive flood.

To the east of Sinya, in the northern Khasi Hills, the Scob
river flows in a narrow gorge of about 2,000 feet in depth, whose
sides are scored throughout its length by innumerable landslips.
Near Sinya village one much larger than the rest came down and,
shooting across the valley, formed a barrier, of which the remains
are more than 200 feet above the present level of the river bed,
Above this barrier, the waters of the Scob accumulated in a great
lake till, about the 7th September,—I was not able to ascertain the
exact date—the barrier burst and a flood of water rushed down
the valley. Three miles lower there is said to have been another

( 221 )

122 OLDHAM: GREAT EARTHQUAKE OF 1897.

landslip as large as that at Sinya, but it was not able to stand
against the rush of the waters, while all the way down the eighteen
miles to where this river issues from the hills, the old channel was
encumbered by innumerable landslips, none able to check the rush
of the flood, which, picking up from them a mass of boulders, gravel,
sand and trunks of trees, carried it along in one great rush,
dealing destruction for many miles within and beyond the limits of
the hills.

Where this river leaves the hills at Ukiam and enters the Kam-

rup District it takes the name of the Kulsi, having been joined by an
affluent from the west, known as the Sri Nadi. At Ukiam it is
joined by a small tributary from the west known as the Dharan,
which drains some low hills by a series of valleys of very low gradi-
ent, the lower part filled with alluvium. Still lower down, the Kulsi
is joined by an affluent known as the Lokia Ján (Khoolsee Khal of the
Revenue Survey maps), a sluggish stream flowing through an alluvial
flat, from which groups of low hills rise.
_ The effect of the flood in this part of the river course was
remarkable and, though not seismic, may be noticed here. The
direct effect of the flood was to sweep everything before it ; trees
were uprooted and soil and subsoil washed away from the underlying
rocks, but in the side channels the rush of the water was upstream.
As a result, I found in the Lokia Jan that trees had been bent over,
and vegetation flattened down in an upstream direction, while every
obstruction had an accumulation of driftwood and weeds on the
downstream side, or just the reverse of what would have been the case
had the flood been an ordinary one coming down this channel. The
distance that the flood rushed up this channel, with sufficient force
to leave conspicuous traces of its effect, was fully two miles, and at the
junction of the tributary with the Kulsi was a great barrier, some
30 feet high and 300 feet across, composed of a matted mass of tree
trunks that had been swept down from the hills and stranded there.

In the Dharan, the same effect could be seen; a short way from
( 122.

\

— EE = e ہیں‎ sE

THE LANDSLIPS. 123

the junction stood a tall fig tree, with a runk of about six feet in
diameter, but rotten at the core. This had been broken off, probably
by the earthquake, and the rush of the flood wave travelling up the
valley had picked up the trunk and carried it full quarter of a mile
upstream.

The indirect effect of the flood has been to raise the bed of the
Kulsi, This can be seen in the main stream, which had formerly a
series of deep pools and was a well known fishing river, but now
flows over a broad and shallow, sandy bed. In the side streams it is
even more.conspicuous, for the raising of the bed of the main stream
has ponded up the water in them, and led to the submergence of a
large area of ground. Inthe Dharan the streams have been flooded
for a distance of five miles and more from the junction, and all the
cultivation of Ukiam village submerged. Close to the junction, 1
found a depth of 20 feet of water in the Dharan, but this is local and
due to the scour of an eddy from the Kulsi; further back the
depth of water was about 12 feet, showing that the bed of the Kulsi
had here been raised not less than 10 feet above its former level,
In the Lokia Ján and another small tributary flooding had also taken
place, but to a less degree. |

In the Sri Nadi, which joins the Kulsi just above the Dharan
there had been similar flooding, but the depth of water was only
about five feet, and only extended a short distance, to a sandy delta
composed of material brought down by the Sri Nadi, which had

nearly joined on to the deposits in main stream. The Sri Nadi

differs from the other tributaries already referred to, not only in its
greater size but in draining the higher hills by a steep sided, narrow
gorge, on the sides of which are many landslips. Evidently the stuff
washed down from these and from the valley of the main stream and
other tributaries joining it below the landslips at Sinya, had already
raised the level of the bed by over five feet, before the great flood
came down; the remaining rise of five feet, being due to this flood
and to debris brought down during the rest of the rainy season.

(^ 12». )

124 OLDHAM: GREAT EARTHQUAKE OF 1897.

CHAPTER VIII.—THE AFTERSHOCKS.

The study of the aftershocks of earthquakes is one to which
much attention has been paid of late years, and the investigations by
Messrs, Omori! and C. Davison? of the records of the aftershocks of
the great Japanese earthquake of 1891 have thrown light on the laws
which govern their gradual diminution in number, and the gradual
shifting of the main centre of seismic activity. -
` In India there is no regular system of recording earthquakes, such
as has been established in Japan, but one of the first steps taken
after the great earthquake of last year was to interest as large a
number of observers as possible in keeping a record of the shocks
felt by them. Such record is of course much less perfect than one
kept by automatic instruments, but in spite of this drawback a large
number of valuable observations have been recorded, which will be
published in a separate part of these Memoirs so far as seems neces-
sary. At present the time has not come, nor the time been avail-
able, for a collation and discussion of these observations, but a brief
review of the general course and character of these series of after-
shocks may be given.

At Shillong, Tura, and all other places within the uer
area, as defined in the next chapter, the shocks felt during the day
following the great earthquake were to be numbered by hundreds.
In fact it is impossible to give a number, for it does not seem to be
an exaggeration to say that for daysthe earth never came to rest, but
was in a constant state of gentle tremor, interrupted at frequent
intervals by distinctly sensible, and again at longer ones by severe
shocks. At the Bordwar tea estate, situated right on one of the focal
fractures which extended to the surface, I was informed that for a week
after the great shock the surface of a glass of water standing on a table

1 On the aftershocks of earthquakes Jour. Coll. Sci. Imp. Univ. Japan, VII, 111-200
(1894).

? On the dístribution in space of the Accessory shocks of the great Japanese earthquake
of 1891 : Quart. Jour. Geol. Soc. LII , 115 (1897).

( 124 )

THE AFTERSHOCKS. 125

was in aconstant state of tremor, and at Tura I was informed that a
hanging lamp was kept constantly on the swing for three or four
days. Some idea of the frequency of the earthquakes in Shillong
may be gathered from the fact that a record kept on the night of
the 19th June, seven days after the earthquake showed 33 distinct
sensible shocks in 44 hours, or an average of one nearly every 8
minutes,

Besides the innumerable smaller shocks there were many which
would have attracted considerable attention in any other years, and
not a few that would have been classed as great earthquakes and
have become the subject of special investigation. Dwarfed as they
were, however, by the great one which preceded them they attracted
comparatively little attention, and it is impossible to get much more
information about them than a somewhat imperfect record of the
area over which they were felt.

One of these, occurred at about 1-30 A.M., local time, of the
13th June, and another about 113 hours later, Both of these were
severe enough to have caused great destruction in the central area,
had there been anything left to destroy, and were felt as far as
Calcutta—the former even as far as Sutna, on the East Indian Rail-
way beyond Allahabad. At 10-40 P.M., of the 13th there was
another shock of sufficient extent to be felt at Calcutta, and again
at 0-47 A.M., on the 14th. After this there was a falling off in
number and violence of the shocks, but on the 22nd June, at 7-24
P.M., and 29th at 10-19 P.M,, shocks were again felt in Calcutta.
Towards the beginning of August there was a revival of seismic
activity, and on the 2nd another shock which was felt at Calcutta at
8-58 P.M. One more was felt on gth October at 1-40 A.M., and since
then there has been no shock severe and extensive enough to be
felt at Caleutta.

In the early days following the great shock, the only detailed
records 1 have been able to obtain refer to earthquakes of considerable
severity and covering a considerable area, but as more complete

( 125 )

126 OLDHAM: GREAT «EARTHQUAKE OF 1897.

records were obtained it was found that those from the central
area were not only remarkable for the number of shocks recorded,
but also for the fact that a very large proportion of these were
distinctly local; shocks being recorded at one station for which no
equivalent could be found ata neighbouring one. Toa large extent
this was doubtless due to inevitable imperfections of the records,
but after making every allowance for this there remained ample
evidence of the local character of many of the shocks, That is to say,
the subsequent earthquakes did not start from one centre, but from a
number of centres scattered over a large area. »

There is one point of view from which these shocks can be
considered, which has an important bearing on the determination of
the nature and extent of the seismic focus. This is the distribution
in space of the shocks felt in the period immediately succeeding
the earthquake.

At Shillong and Tura they were, as has already been remarked,
to be numbered by hundreds a day for the first few days, At Kau-
nia the Station Master reports 41 shocks after the great one, up to the
end of 15th June, At Maimansingh the meteorological reporter
counted 83, while in north Gauhati, there were no less than 561 in
the same period according to a correspondent of the Assam
newspaper : |

From 15th June to 15th July I have records from Kaunia,
Kuch Bihar and Maimansingh, and from north Gauhati, on the
authority of the correspondent of the Assam referred to above;
the number of shocks are as follows :

15 to 30 June, I to 15 July.
N. Gauhati . ` : 125 84
Kaunia . A o 57 sus
Kuch Bihar . e . 64 31
Maimansingb. . e 52 40

These show that the shocks were much more frequent in north
Gauhati than would be expected if the centre of the earthquake
was wholly within the hills south of the Brahmaputra Valley, `

( 126 )

THE AFTERSHOCKS. 127

From the latter end of July I have a valuable set of records of
earthquakes, kept by the hospital assistants in charge of the dispen-
sary in the Goalpara District, for which l am indebted to Surgeon-
Major E. F. H. Dobson, Civil Surgeon of the District. The value of
these varies greatly and at some stations the record is obviously in-
complete, but the broad lines of the relative frequency of earthquakes
may be accepted.

The first point to notice is that they were very frequent at all
places near the north foot of the Garo Hills. Thus for the period 1st
to 15th August the record for Tura being 124 shocks there are

reported—
151 shocks from Darangiri,
182 "y Aa Goalpara,
6۶ » Lakhipur,
OR 5 h Krishnai,

while the total number of shocks reported from Dhubri, as the
combined record of several observers is only 48 for the same period,
and if we go still further west, we have only 12 shocks reported at
Kuch Bihar and 28 at Rangpur.

Further north in the alluvial plain we have the town of Bijni in
the extreme north of the Goalpara district, where no less than 105
shocks were reported by the hospital assistant, while from Rupsi
10 miles to the N. W. of Dhubri, only 10 are reported for the same
period Ist to 15th August, The Rupsi record is probably incomplete,
but still there is good reason for concluding that earthquakes were
really much less frequent there than at Bijni, and this is confirmed
by the reports from Borpeta in the Kamrup district, whence 113
distinct shocks were reported in the period 1st to gth August.

Apart from these records, which might be considered to be taint-
ed with the inaccuracy attaching to most Indian statistics, I have
independent and unquestionable evidence that long after earthquakes
had become rare at Gauhati they were still being felt at the rate cf
several a day at Borpe*a.

( 127 )

128 OLDHAM: GREAT EARTHQUAKE OF 1897.

It is to be noticed that most of these earthquakes were feeble
ones, which were not felt over any large area, and were forthe most
part local. We have consequently an area covering the part of the
alluvial valley of Assam on either side of the g1st meridian, where the
conditions were very similar to those of the central area in the hills;
that is, it was an area where small shocks were frequent and much
more numerous than in the country outside.

If we turn to the south of the hills we find a great contrast. Not
only are the reported shocks much fewer, the combined records of
Sylhet and Sonamganj giving only 20 shocks from ist to 15th
August, but there is noindication of their being markedly more
numerous at one place than another in an east or westerly
direction.

To the west of the Garo Hills earthquakes are also relatively
numerous and, what is more important, there is a strip of country run-
ning out into Bengal where the conditions are very like those of the
central tract. This is well shown by the records from Rangpur, which
appears to have felt quite a number of small shocks that were
not recorded at the neighbouring stations, from which regular returns
were being received. ' i

An examination of the area over which many of the larger shocks
were felt, shows that in most cases the shocks that were felt over
the Assam Valley and north-eastern Bengal did not penetrate
into Sylhet and Cachar or were only slightly felt at places close
to their northern boundary. ۱

Taking these facts into consideration we see that the centres of
the aftershocks, in the period shortly following the main shock, were
not symmetrically situated with reference to the axis of the range
separating the Brahmaputra and Barak Valleys, but that they lay for
the most part towards the northern edge of these hills or under the
alluvium to the north over an area which extends northwards along
and near the gtst meridian of East Longitude.

( 128 )

THE EPICENTRAL TRACT. 129

CHAPTER IX.—RESULTS OF AN EXPLORATION
OF THE EPICENTRAL TRACT.

During the fine weather of 1897-98 I had an opportunity of
making an examination of the epicentral tract, which could not,
unfortunately, be so complete as, for instance, that undertaken by
the United States Geological Survey in the case of the Charleston
Earthquake of 1886.

The reasons for this are three-fold: firstly, the great area;
secondly, the impassable character of much of the country ; and
thirdly, the limited time available. Of these, the last may be
regarded as the corollary of the first ; had the area to be examined
been only as great as in the case of the Charleston Earthquake, the
time would have been ample, but instead of an area with the largest
dimension of about 30 miles, the corresponding tract of ground
in the present case had a length of about 300. Added to the
size of the area was the fact that a considerable proportion of it, and
that too where the most extensive permanent changes were produced,
is covered by a dense growth of forest or bamboos, through which it
is only possible to advance by laboriously cutting every step of the

way, once the beaten track is left. For this reason the examination
bad to be practically confined to the beaten tracks, which were only

left when there seemed good reason to suppose that the results
would repay the time taken up, in slowly forcing a way through the
jungle at a rate of about three hours to the mile,

In describing the results obtained I do not propose to adopta
narrative form. On the map No. II is shown the course actually
followed on this tour of examination, and many of the results ob-
tained have already been incorporated in other chapters of this re-
port. The present one will be chiefly confined to a consideration
of the evidences of permanent changes of level due to, or perhaps
rather the cause of, the earthquake.

Before dealing with these, it may be well here to notice the
K | | ( 129 )

130 OLDHAM: GREAT EARTHQUAKE OF 1897.

evidences of a large vertical component of the wave motion which
were to be observed throughout the whole of the area examined.

In Captain Howell's report on the earthquake in the Garo Hills
it is mentioned that a stone lying on the surface of the parade
ground was thrown a distance of three feet, and as, from the de-
scription, it is evident that the stone had travelled through the
air, and not along the ground, from its original to its subsequent
position, itis evident that the impulse given to it must have been
upwards as well as sideways.

Many other cases, similar to this, were observed in other parts of
the Garo Hills, but they were not so conspicuous asthey might
otherwise have been, owing to the high and rank vegetation which
covers these hills. It is different on the bare plateau of the Khasi
Hills, where the combined effects of greater altitude, smaller rainfall
and annual fires have denuded the greater part of the hillsides of
all vegetation except grass. Here we have grassy slopes on which
numerous rounded blocks of weathered crystalline gneiss or granite
lie more orless imbedded. Of these, many were driven from their
seats and hurled through the air, leaving a sharply cut mould in the
soil, slightly broken down on the side towards which the block was
projected. These moulds are commonly seen on slopes, as those
in Plate XXXI, but are also to be found on level patches of
ground, while the fact that the stones were actually projected through
the air is shown by the marks of the spot where they fell, at varying
distances from their original position, and the absence of any trace
in between of their having touched the ground. At Nongstoin I
noticed that a piece of granite, about 3 ft. long by 1 ft. broad and 9
inches thick, which had been lying flat on the ground, had been
thrown upwards into the air to such a height that it broke across
in falling. |

Between Mokersa and Nongstoin, I measured the distances to
which several of these blocks had been thrown on level ground, and
found it varied from 2 ft. to 4 ít. as a rule, and these may be taken

( 130 )

THE EPICENTRAL TRACT. j 151

as the general limits throughout the tract examined, When the
force of the earthquake was sufficient to tear these blocks from their
site, it appears generally to have been sufficient to project them
about 2 ft., and though smaller distances of projection are not uncom-
mon, they are fewer in comparison than the greater ones.

About 4 ft. is the usual upper limit, though this is exceeded locally,
The greatest projection, and the most marked evidence of vertical
movement, [ saw was about four or five miles north of Rambrai,
where a long splinter of granite, 3 ft. long and of triangular section
with sides of about 12, 10, and 9 inches,—a naturally weathered
fragment which had been lying flat on the surface of the ground—
was thrown 8ift. from its original position. About quarter of a
mile from this a group of small Khasia monoliths, some 6 ft. high,

Fig. 14. Disturbance of Khasia monuments near Rambrai.

had been destroyed, not by breaking off, nor by upsetting, but by
being shot upwards out of the ground as indicated in fig. 14. One
of these had travelled 6} ft. through the air from the place where
it had stood tothe place where a deep dent marked the spot at
which its lower end had first struck ground. The socket in which its
base had been buried was almost uninjured, and the angle at which
it was shot upwards could not have been much, if at all, less than 60?
with the horizon.

It may be noticed that in almost every case of projection of

'ablock of stone it also turned over in its course, and now lies with

its originaly lower side upwards, This overturning of objects
K 2 ( 131 )

132 OLDHAM: GREAT EARTHQUAKE OF 1897.

projected by earthquakes is a commonly observed phenomenon, and
may in part be attributed tothe nature of the impulse imparted by
the earthquake wave. In the case of these projected stones, how-
ever, the explanation is probably more simple. In every case
wherea stone, partially imbedded in the earth, was shot out in a
slanting direction, it had to break down one edge of the cup in which
it lay, and even where, owing to the form of its under surface no
actual breaking away of the earth was necessary, there was a
greater frictional resistance than on the opposite side. Consequently
that side of the stone, which lay towards the direction in which it was
thrown, had a slight drag put on it, and soa spin was imparted to the
stone, which made it turn over in its course.

Another point to be noticed is that the stones which were
projected were all much ofa size and, with very few exceptions, from
about 1 ft. to 3 ft. in diameter. The reason of this is not far to seek ;
each stone, partially imbedded inthe ground, was restrained by the
adhesion of the soil to the stone, a restraint which had to be overcome
by the momentum imparted to the stone by the earthquake wave.

The adhesion is proportional to the surface in contact with
the earth, and increases with the square of the dimensions of the
stone ; the force which tends to overcome this adhesion varies with
the mass of the stone, that is to say, with the cube of the dimensions.
In the case of very small stones, then, the momentum imparted to the
stone is not sufficient to overcome the attachment, but as the
dimensions increase a limit is reached, when the pull of the stone is
sufficient to allow it to become detached from the soil. The exact
size of stone at which this becomes possible depends on several
factors, the maximum velocity imparted to the stone, its specific
gravity, the nature and degree of moisture of the soil, and the form of
the cavity left by the stone. With the conditions prevailing in the
Khasi Hills, at the time of this earthquake, the limit was, as has been
stated, about 1 ft. in diameter. Smaller pieces were occasionally
projected, but this was comparatively rare, except in the case of
pieces lying absolutely loose on the surface.

( 132 )

THE EPICENTRAL TRACT. 133

The upper limit of size is due to another cause. The motion of
the earthquake wave has to be communicated to the stone through
the, comparatively, yielding and inelastic soil, in which it is partially
imbedded. But the stone does not take up the motion immediately
or without some resistance from its own inertia, which increases
` with the mass of the stone, and part of the energy of the shock is
consequently taken up in compressing the soil under the stone. Inthe
case of smaller fragments this proportion may be small, the stone
quickly takes up the movement of the earthquake wave, and the
velocity communicated to it may be not much less than that of the
wave particle in the rock below the layer of soil. The larger the
mass of the stone, however, the slower it takes up the motion of the
wave and the greater the work expended in compressing the yield-
ing soil beneath it; after this deduction, the velocity of movement
actually imparted to the stone may be insufficient to enable it to
leave its bed and travel as a projectile through the air.

This upper limit was about 3 ft. in diameter, the fragments being
generally of a rudely spheroidal form. Larger rocks were displaced in
abundance, dislodged from their sites, and caused to roll down the hill-
sides, but I saw very few cases of stones as large as 3 ft. in diameter
which had been actually projected in a free course through the air.

Both at Goalpara and Gauhati instances of objects projected
through the air were observed, and at the latter place there is inde-
pendent evidence of a high angle of emergence. In the centre of
the river, opposite. Gauhati, rises a small island, commonly known
as Peacock Island, which is regarded as sacred, and crowned with an
ancient temple. It is stated that before the shock of 1897, though
earthquakes had visited and damaged both banks of the river, they
had never been able to reach this island, an immunity which was
attributed to its sanctity, but finds a rational explanation in the deep
channel of the Brahmaputra which surrounds it,

The investigations in Japan! have shown how completely the

1 J. Milne : Trans. Seismol. Soc., Japan, VIII, 98; S. Sekiya and P, Omori, Jour. Coll,
Sci, Imp. Univ. Japan, 1V, Pt. ii, 249 (1891).
| ( 133 )

134 OLDHAM: GREAT EARTHQUAKE OF 1897.

destructive earthquake waves are confined to the surface, and how a
trench of only 20 ft. deep is sufficient to cut them off and protect a
building by which it is surrounded. In the case of Peacock Island,
the channel of the Brahmaputra served asa protection against
earthquakes, at any rate against those which originated at a distance
and were travelling nearly horizontally. The surface waves were
exhausted against the edge of the river bank, and the main wave,
passing under the channel of the Brahmaputra, left the island, as it
were, in a shadow and protected from anything but an insignificant
tremor. |

The great earthquake of 1897, however, instead of reaching
Gauhati from a distance, was travelling upwards from below. The
bed of the Brahmaputra consequently gave no protection, and the
temples on Peacock Island suffered as severely as those on either
bank of the river.

Before proceeding toa consideration of the permanent changes,
which accompanied this earthquake, it will be well to describe briefly
the physical geography and structure of the country involved. The
group of hills lying south of the Brahmaputra Valley js com-
monly known in different parts by names derived from the tribes in-
habiting it, but no general name has found its way on to maps.
In previous publications of this Survey the names of Shillong plateau
and Assam range have been proposed and used, and of these the
former is the most descriptive of its character, the latter most con-
venient for general use. It is an elevated tract composed of cry-
stalline gneissic and granitic rocks, with some metamorphic schists
and quartzite, which carries a varying thickness of cretaceous and
tertiary rocks along its southern edge. These newer beds thin out
to the north, not by denudation but by an original thinning out
due to deposition on a sloping sea bottom. For about 50 miles,
along the north of the Sylhet plain, this hilly region ends abruptly
in a uniclinal flexure or as faulted scarp. At its western end

( 134)

THE EPICENTRAL TRACT. 135

there is a fringe of low hills between the edge of the highlands
and the plains of Maimansingh. To the east, the line of flexure
and faulting bends round to the north-east and appears to become
one of the outlying members of the Patkoi or Bare! system of disturb-
ance; but too little is known of the geology of these parts to
assert this with certainty.

Viewed from one of the higher points the plateau character of
this mass of mountains is conspicuous, From Mao-phlang, for instance,
the view is one of low, round topped. ridges, all rising to the same
general level and separated by shallow, open valleys, generally
having a flat, peaty bottom. The character of the country is that of
one which, to use a term for which we are indebted to the Ameri-
can geologists, has been ‘ base-levelled,’ that is to say, has been worn
down by the action of subaerial denudation till rain and river can
have but little further effect in altering its form. Such a form of
surface could never have originated on these hills in their present
form ; it points to a much lower elevation, and a change in the charac-
ter of the hills is even now going on, for everywhere deep, steep sided
gorges have been driven into the elevated mass from the lowlands
on either side, The contrast between these deep gorges and the
open, shallow valleys of the uplands is striking, and what is im»
portant to note, as showing the geologically recent date of the
upheaval, the transition from one type of valley to another is abrupt.
The deep gorges run up into the hill mass, maintaining their
depth for a varying, but generally considerable, distance ; then
comes a comparatively short stretch of steep gradient, waterfalls
and cascades, and the valley passes abruptly into the open base-
levelled type of the highland plateau, which has as yet been

uninfluenced by the altered conditions consequent on the change of
elevation,

The deep gorges are not, however, the only places where steep
and high slopes are seen. Traversing the plateau are a number of
what, seen from one side, appear to be high ranges, all of which

(135)

136 OLDHAM: GREAT EARTHQUAKE OF 1897.

agree in the character that the crest of the range is the edge of
another plateau sloping away from the summit of the steep ascent.
These ranges vary in length and height, for the most part they
have a general east and west or east-south-east and west-north-west
direction, and die out gradually at-either end, though occasionally `
they end up in a cross range or scarp.

Of this character are the Tura and Arbela ranges in the Garo
hills, shown in fig. 15; the high ridge of the Shillong peak, which over-

Fig. 15. View of the Arbela range, from the road to Tura.

looks the lower plateau on which the station of Shillong is built is of
a similar character, as also the Maothadraishan (Mauterichan) range,
along whose foot the road to Nongstoin runs, and many other
ranges. Looking from the summit of any of these ranges it is im-
possible to doubt that the undulating plateau which stretches away in

(- 3309)

THE EPICENTRAL TRACT. 137

one direction from the crest, was once continuous with the precisely
similar plateau which slopes up from the foot of the range on the
other direction. These ranges are in fact fault scarps, due to unequal
elevation along faults, by which the rocks composing the range have
been split.

The history of the Assam range, consequently, falls into three
stages :—First, an old land surface which had long been exposed to
denudation and worn down till the agencies of subaerial denudation
had almost ceased to produce any further change, and the land
surface was reduced to that aspect for which Prof. W. M, Davis has
proposed the name ' peneplain.’ Then this land surface was elevated,
and the elevation was unequal, the rocks were broken through by
fissures of which one side was raised further than the other, and instead
of a uniform surface the old peneplain was broken up into portions
which lay at different heights. Finally, the elevation of the plateau as
a whole gave the streams a chance of eroding once more, deep
gorges were driven into the range, and in parts the re-shaping of the
surface has gone so far that almost all trace of the old peneplain is
lost.

One point more may be noticed, that, so far as the main mass
of these hills is concerned, the last elevation does not seem to
have been accompanied by any marked compression. Undulations
of the plateau, sometimes of considerable height, have been formed,
besides the fault scarps, and these may indicate some compression.
It is possible, too, that the faults are reversed ones, and so in-
dicate compression to a small degree, but on this point there is no
evidence available. Any way, the amount of compression which has
taken place, if any, must be small in the main mass of the range.
On the southern edge, where we come into the region of tertiary and
cretaceous sedimentary rocks we also enter a region where there has
been a certain amount of compression, as evidenced by the folding
of the strata. Though the amount of compression undergone is
small in proportion to that which the rocks of the Himalayas have
undergone, it is not inconsiderable and points to an approach

(137)

138 OLDHAM: GREAT EARTHQUAKE OF 1897.

between the main mass of the range to the north and the portion of
the earth’s crust which lies to the south, The consideration of the
nature and direction of the movement which has caused this must,
however, be deferred.

After this brief account of the physical geography and structure
of the country concerned, we may proceed to a consideration of the
evidence of permanent changes due to the earthquake. These
may be classified as follows :—(1) Faults and fractures, (2) differ-
ential changes of level, evidenced by interruptions of drainage, un-
accompanied by faulting, (3) changes of level evidenced by reported
changesin the aspect of the landscape, and (4) changes proved by
a re-observation of the triangles of the Great Trigonometrical Survey.!

Of the faults and fractures, the most conspicuous example is
that of the Chedrang fault, as I shall call it, from the name of the
stream whose valley it follows. A sketch plan of part of the course
of this fault, showing its effect on the drainage courses it crosses
is given on Plate XLII, and in describing it, I propose to start at
the point where it dies out, or at any rate beyond which it was not
traceable, some six or seven miles SSE from where the Chedrang
enters into the plains at Dilma.

At this point the fault is not directly recognisable by any
throw at the surface, but the hillside is very much fissured, the
vegetation has been much disturbed, and not a few small trees
have been snapped off like that depicted in fig. 16, p. 150.
Following the line of this disturbance to the NNW the fault
crosses a small tributary, with a throw of 2 feet, and shortly after `
crosses the Chedrang river diagonally, On the upstream side of
the fault a depression has been formed which was almost filled up
by a deposit of sand.

Quarter of a mile down stream the fault, still running about
NNW crosses the stream once more, and has here a throw of about

1 The results of this re-observation were not received in time for incorporation in the body
of this report. A detailed description will be found in Appendix G.

( 138 )

THE EPICENTRAL TRACT. 139

2 feet, the upthrow being, as everywhere else, on the eastern side.
After continuing on the east of the river for about half a mile, the
fault crosses it once more and, the upthrow being on the down stream
side of the fault, the river has been ponded up for a distance of
about quarter of a mile upstream. At this point the fault is double,
being two parallel step faults separated by about 20 yards.

Continuing to NNW the fault keeps in the valley, close to the
main stream, and has given rise to a number of small pools by
blocking the side drainage. One of these, which had no outlet, was
about six or eight feet deep, to judge by the appearance of the
drowned vegetation in it ; the bed of the old watercourse was re-
cognisable, and its vertical height above the surface of the water
was 20 feet, as determined by an Abney’s level. As this old stream
bed was formerly continuous with the one now submerged in the
pool, the total throw of the fault cannot here be less than 25 feet.

Immediately beyond this pool the fault once more crosses the
Chedrang and forms the waterfall shown in Plate XIV. The waterfall
is a direct result of the fault, but it does not occupy the exact fault
line ; this is seen in section on the river side, and the rock on the
east is seen to be shattered by a great number of nearly parallel
and. vertical fissures, which cut the rock into small fragments. These
have been washed away by the rush of the stream when in flood, for
a distance of about 20 feet back from the fault line, and the frag-
ments dislodged have accummulated below the fall, reducing its
height to about 9 feet. `

Below the waterfall the fault runs along the stream bed for
about half a mile, and the river now flows in a depression on the
downthrow side of the fault, while the old river bed lies high
and dry, as shown in Plate XV. Here we see on the right the old
river bed with its waterworn boulders ; on the left is the present
channel, and separating the two is the fault scarp, composed in the
foreground of rock, and further back of river boulders from top to
bottom.

About half a mile below the waterfall, the fault leaves the river

( 139 )

140 OLDHAM: GREAT EARTHQUAKE OF 1897.

bed and runs for a short distance close to, but west of, the river.

before it enters a broad sheet of water. This pool, whose recent
origin is shown by the number of dead trees and bamboo clumps
standing in the water, has a maximum depth of 18 feet ; it covers the
course of the fault, but the inequality.of ground level could be traced,
by sounding, for some distance into the pool. About quarter of a
mile from where it enters the pool, the fault leaves it once more, and
here the throw has been reduced to almost nothing. Following the
line of fault, it is found to increase in throw and to form a large pool
by interrupting the drainage of a tributary valley. The throw
increases, in about quarter of a mile, from nothing to 25 feet and
a short way on, where the fault rises up ontothe hillside, to31 feet.

Returning to the pool mentioned above, through which the fault

runs, it has a length of fully half a mile and a breadth of between
300 and 400 yards. The noteworthy point about it, however, is
that it differs from the other pools noticed as yet, in the fact that its
barrier is not directly formed by the fault. The pool spreads across
the fault, and its outlet in the old channel of the Chedrang lies to the
east of the fault line. At the outlet there is nothing in the shape of
a barrier visible, the channel of the stream merely gets deeper in an
upstream direction, and its bed sinks gradually, with no sudden
change, under the waters of the lake. In other words, the natural
slope of the channel has been reversed and a depression formed in
which the water accummulated. Now, if we return to a consideration
of the fault, we see that the maximum width and depth of this
pool is where the fault has no throw, while the barrier corresponds to
a part of the fault where the throw is rapidly increasing. Hence, we
see that this pool, or lake, is not directly due to the fault itself, but to
the formation of a roll or undulation in the surface of the ground to
the east of, and corresponding in position with, an increase in the
throw, of the fault.

Returning once more to the place where the throw of the fault
was 31 feet, and where, having temporarily left the valley, it was

running up the side of aspur. The spot is close to, but a few

( 140 )

THE EPICENTRAL TRACT. 141

hundred yards up stream from, where the Chedrang is joined by
a considerable tributary from the east, just where the village of
Dalbot is marked on sheet 124 of the Atlas of India. At this
place the destructive energy of the earthquake is very marked.
Huge blocks of granite, ten and twenty feet in diameter, have been
dislodged from their positions and thrown about. One block,
roughly cubical and about 40 ft. x 30 ft. x 30 ft., happened to lie across
the line of fault and was overturned by it, Clumps of bamboos were
bodily overturned, and bamboos and trees, even where still standing,
were killed, evidently by the violence of the shock; in sucha
climate it is impossible to attribute their death merely to a loosening
of the soil and consequent drving up of the roots.

Near the junction of the tributary referred to above, and quite
200 feet or more in a direct line from the main fault, the gneiss in
the Chedrang river bed was seen to have undergone some disturbance,
and along numerous planes of jointing, or old slickensides, a small
movement of a few inches had taken place, always with the upthrow
on the east.

Proceeding down the main valley, the fault is struck once more,
after having crossed a spur of the hills to the west of the Chedrang,
and crosses the river obliquely, forming a waterfall. The throw here
has sunk to about 8 feet, the upthrow being still on the east and the
general course about north-north-west, with numerous local irregu-
larities. The fault keeps close to the river for about half a mile
crossing it three times and forming a pool when the upthrow is on the
downstream side, or a waterfall where it is on the upstream side;
the throw of the fault increases to 18 feet measured, and probably
more where there were no means of exact measurement.

The fault has now crossed to the west of the river and continues
so for nearlya mile. The throw dies out till it is only traceable with
great difficulty, but increases once more, and the fault crosses a series
of low spurs, forming three lakelets by blocking the side drainage.
Two of these are of fair size, of triangular shape, one about 300 yards
in length along the barrier formed by the fault, the other, situated

(E141)

142 OLDHAM: GREAT EARTHQUAKE OF 1897.

close to where the village of Dilma is marked on the Garo Hills
District map and on sheet 124 of the Atlas of India, is rather larger.
The fault here has a throw of about 20 feet, and in this lake two
Garos are said to have been drowned. According to the account
given me, they were travelling along a footpath, which ran down this
side valley, when the earthquake took place and the waters collected
with such rapidity that they were drowned. If there is any truth at
all in the story, it is probable that they found themselves close to the
fault line and were stunned either by the violence of the shock or by
a falling tree or rock.

Returning to the place where the throw of the fault had fallen
to nothing, it must be noted that here againa large pool is formed
in the main stream, but altogether east of the fault. Owing to the
valley being more open, filled with alluvium, and the gradient
lower, this pool is larger than the similar one further up stream. It
was more than half a mile long and nearly as broad at its broadest
point, even when I saw it in February, and had evidently extended
even further on either side, over the almost level alluvial plain,
before the outlet was cut down. Its outlet is similar to that of the
upper pool, except that instead of a stony bed there is sandy alluvium ;
the bed simply sinks under the water of the lake, and the change
from lake to stream is gradual and unaccompanied by any faulting or
sudden break. |

Returning to the fault where we leftit, it very soon crosses the
Chedrang, but the bottom of the valley, being here filled with
alluvium instead of having a rocky bottom, no waterfall has been .
formed. The alluvium has, however, been greatly eroded and a
great sandy fan or delta formed immediately down stream.

The fault scarp can be easily traced in the alluvium to the
east of the river bed and, about a mile below where it crosses the
river, had a measured throw of 32 feet. From here on the fault en-
ters a region where the alluvium is thick, and no longer shows asa
broken scarp, but as a short slope. Where this runs through forest
the trees have been tilted over in the remarkable manner seen in

( 142 )

+

THE EPICENTRAL TRACT. 143

Plate XVI, and at the outer edge of the hills, near Kacharipara of
the Survey of India maps, it runs through the rice fields, which
have been tilted up in a smooth, unbroken slope between the upper
and lower levels on either side of the fault.

Here the fault runs out into the open plain, where the thick-
ness of alluvium is very great, and consequently cannot be followed
with definiteness, but it does not die out, for its effects, as those
of another fault on about the same run, are very noteworthy at
Jhira.

The village of Jhira, which owed what importance it had toa
weekly fair, was situated on the left bank of the Krishnai, about
three miles below the confiuence of the Chedrang. The bazar is
now flooded, and in February there was about three feet depth of
water over it. Northwards from Jhira, and extending to the foot
of the hills on the west, there is a broad sheet of water, about a mile
and a half long, about three quarters of a mile broad, and some 12
feet deep. At Jhira, the river bed is obliterated by the waters
which stand above its banks; following the old channel down
stream the banks gradually rise out of the water, which gets shallower
till finally the river bed risesin a gentle slope of dry sand. Ac-
cording to the reports of the engineer who was deputed by the
Assam Government to enquire into the floods consequent on the
earthquake, this barrier rises to a height of about 15 feet above the
present dry weather level of the Krishnai and has a length of a
quarter of a mile to where the bed again carries water, brought into
it by a tributary.

On the east side of the Jhira lake there is ample evidence of
change of level, for part of the dry land was formerly fhil and per,
petually under water, and at one place the remains of an old irri-
gation channel can be seen, which formerly carried water to the
low lands on the east, but now rises steeply from the water level.
At the northern end of the lake the drainage now makes its escape
in a broad and shallow sheet of water over what was once high
land covered with sa/ forest. After a short course of this nature it

( 143 )

144 OLDHAM: GREAT EARTHQUAKE OF 1897.

has found an old drainage channel, of a small tributary, and widened
this out, falling into the old course, after a succession of rapids,

about a mile and a half from the northern end of the lake, or about

three miles in a direct line from Jhira. Between the lake and the

junction of the new and old courses, and between them and the low
spur to the west, the flood waters have spread over a large area of
what was once high land, and are estimated to have killed not
less than 50,000 sal trees.

The barrier by which this lake is formed, that which crossesthe main
channel of the Krishnai, appears to be directly due to the fault, which,
on account of the great thickness of alluvium, manifests itself as a
gentle roll or undulation of the surface and not as a sharply defined
scarp. Following the line of the fault it should strike a long spur
of low hills which runs out northwards into the alluvium, but I could
fnd no trace of it there, and the fault appears to have died out.
Whether it re-appears further on along this line I cannot say, as
] had no opportunity of exploring this country. The water, however,
now finds its escape over ground which formerly lay at a higher level
than the stream bed, and this is now lower than the crest of the
barrier across the old channel, though lying on the upthrow side
From these facts we may conclude that the throw of the
fault must have become less, inthe neighbourhood of the present out»,

let of the lake, than at the place where it crosses the old bed of the
that it diminishes to the northwards of
t may die out altogether before

of the fault.

Krishnai; in other words,
Jhira and it is not improbable that 1
e low hills to the north and west of that place.
wards of Jhira, the flooded ground ex-
ne of the bends of the Krishnai, where
the water

reaching th
Upstream, or to the south

tends for about 4 miles, and o
it takes a curve tothe east, has been raised and laid dry,

now finding its course along a depression through what was once forest
land. The country along the course of the Chedrang, from where it
leaves the hills to where it joins the Krishnai, has also been extensive-
ly flooded, in fact the levels of the whole of the triangular patch west
of a line joining the issue of the Chedrang fromthe hills and Jhira have

( 144 )

THE EPICENTRAL TRACT. 145

been altered, Large areas have been permanently flooded, but the
boundaries of the flooded areas are irregular and principally deter-
mined by the original inequalities of the surface of the ground,

Having described the course of this fault in detail there are
certain aspects of it, as a whole, which require notice. Firstly,
although the throw of the fault varies from over 35 feet to nothing, yet,
wherever there is any perceptible throw, the upth»ow is always on
the east and the downthrow on the west, Secondly, I was unable
to detect any pronounced horizontal movement of one side with
reference to the other, in other words, the displacement appears
tohave been simply up ordown, so that the fault is a fault pure
and simple, and not a heave, Thirdly, wherever the plane of the
fault could be seen in rock it was practically vertical, with no pros
nounced hade in either direction. Fourthly, the displacement appears
to have been principally, if not entirely, an elevation of the upthrow
side and not a depression of the downthrow side of the fault.

With reg-:d to the last mentioned point it is not possible to
speak with the same positiveness as on the first three, for the only
measurements that can be made with certainty are differential ones;
that is, the present difference of level of two points which were
formerly on the same level can be measured, but it is not possible to
determine whether one has gone up or the other down. There are,
however, some considerations which bear out the conclusion for-
mulated above, The twolarge pools, which correspond in position
to what may be called the nodes of the fault, that is, the places
where the throw becomes nil, are, as has been explained, due to
the raising of a barrier across the river channel, whereby it was
given a reverse slope. Now it is conceivable that this might be
due to a general and uniform subsidence on the downthrow side,
and an unequal subsidence on the upthrow side, whereby certain

parts retained their original level while in others the subsidence

was equal to that of the downthrow side of the fault. In this way

hollows would be formed on the upthrow side in which water
would accumulate, |

L ER 0

- 4
جج و کے‎ Au o d f CMM P

TT “eri

i 40 OLDHAM: GREAT EARTHQUAKE OF 1897.

This explanation seems to be excluded in the case of the
lowermost of the pools, that at Jhira. Here the rapids between the
present outlet of the lake and the point where its waters join the old
channel, are of the character of those in a mountain stream, although
it flows over alluvium instead of rock. The steepness of the gradient
here could be explained by the raising of the water level due to actual
elevation of a barrier or to a depression of the old stream bed where
it is rejoined by the water. But though, so far asthe short reach of
rapid current is concerned, we might attribute the gradient equally
to a raising of the upper end or a depression of the lower, yet the
latter explanation is here excluded by other considerations.

The whole of the alluvial plain of Lower Assam is raised but little
above the sea, and the gradient of the rivers is about as small as it
could be, so that if, in this area, there had been subsidence, sufficient
to account for the gradient of the stream bed, the area depressed
could not but be flooded with water. Such depressions have, perhaps,
been formed, as will be noticed further on, but so far as I could
discover, nothing of the sort has taken place along the Krishnai in
that part of its course immediately below the Jhira lake.

For this reason we may conclude that, in the alluvium, there
was an actual elevation of the upthrow side of the fault, and by
analogy we may extend the same conclusion to that portion of the
fault which lies within the hills. Atthe same time it is possible that
there may have been more or less depression on the downthrow side,

and that the visible throw at any point is made up of the sum of.

elevation on the one side and depression on the other.

We have now finished with the description of this fault, which has
a visible throw of 35 feet and more in places, and has been followed
fora distance of about 12 miles, Whether this represents its total
length is not, however, certain, for in the one direction it is lost in
the great thickness ofthe alluvium of the Brahmaputra, and in the
other it dies out in thick jungle. In this direction its utmost limit
cannot, however, have been as much as 5 miles beyond the
furthest point to which it was traced, or it would have been noticed

Më Lee

THE EPICENTRAL TRACT, 029

on the route traversed along the head of the Chedrang drainage area
Taking the limit of 12 miles for its length, this, combined with the,
observed throw, would be sufficient to account for a very severe
earthquake, and it is fortunate that it was not discovered at the out-
set of the examination, for it might well have been regarded as a
sufficient cause, and the area of the examination much restricted. As
a matter of fact nearly all of the observations which follow were
made before this great fault was found, and the general result is
| that the Chedrang fault, though by far the most conspicuous, is but a
small fraction of the total of the permanent changes which accom-
panied the earthquake, It must consequently be regarded as only
part of the cause ofthe earthquake,

About 10 miles due south of the termination of the Chedrang
fault is the village of Sámin, past which the main road runs south-
wards from Damra through the Garo Hills. Just before the road
reaches the Ronghri River there is an abrupt rise of about ro feet,
and on either side of this a low but steep rise, in the surface of
the alluvial gravels, runs in a WNW—ESE direction. At first
sight there is nothing to distinguish this from the natural termina-
tion of a terrace of valley gravels, except the steepness and bioken-
character of the road where it drops from the upper to the lower
level. On the left of the road, however, there are as shown in
Plate XIII the remains of an old Garo house. The low cliff just
mentioned runs through the centre of this, and one of the posts, which
supported the raised platform of the house, can be seen standing
below, and another above, the break in the ground level. As the tops
of these posts must have been on a level with each other when the
house was built and as the top of the one is now 10 ft. higher than
that of the other, they show that the throw of the fault, for such it
is, was 10 feet at this point, |

This Sámin fault is smaller, both as regards throw and length,
than the Chedrang fault; its general course is also different, being
about E 30° S—W 30° N and the upthrow is on the southern

SE ( 147 )

145 OLDHAM : GREAT EARTHQUAKE OF 1897.

side as indicated in the plan drawn on Plate XLII, To the east -
of the road it crosses a tributary of the Ronghri, down whose valley |
the road runs, and has ponded up the water for a length of about
quarter of a mile. The fault then crosses a tributary of this stream
and, running up over the hills, forms a small lake, whose old outlet is
now dry,a new outlet having been formed along the fault line ; the
fault runs on over a spur and once more crossing the tributary
stream has formed a small lake. It can be traced for about a mile
further, following the lines of minor drainage, till it dies out and is
lost. In the continuation of the run of the fault in this direction
some small pools were seen to have been formed, but the fault could
not be traced as such.

In the opposite direction it can be traced for rather more than
a mile, forming several small pools in_its course, and is lost in a
broad, grassy plain over which the flood waters of a stream, whose
channel had been dammed by the fault, found their way. The total
length of this fault is thus about 24 miles, with a maximum throw of
10 feet, dying off in either direction to nothing, .

Leaving for the present the mention of a few smaller faults which
were observed, we may proceed to the consideration of the Bordwar
fault-fissure, or fracture. I use this term to distinguish it from the
faults on the one hand, in which there has been a change in the.
relative levels of the ground on the one hand, and on the other hand
from the earth fissures described in chapter VII, which are the
results of the earthquake wave, being in no way connected with its.
cause. The features to which I give the distinctive name of fracture
are of the same nature as: those along which faults are formed,
differing only in the fact that there has been neither up-and-down
or sidelong movement along them. Like the faults proper they are
directly connected with the cause of the earthquake and are foci of
unusual violence of shock,

The largest of those that I saw was what I shall call the Bordwar
fracture, from the name of the tea garden through which it runs,

( 148 )

e

THE EPICENTRAL TRACT. 149

To the south-west of the old bungalow there rises a low hill of
gneiss which was reported to have been rent from top to bottom by
the earthquake. On examination I found that the statement was
correct, though based on an erroneous inference; a line of fis-
suring can be traced up the north-east face of the hills and across
the crest, but it is not the fracture itself which is visible from a dis-
tance and led to the report that was made,

The actual fracture is only a few inches wide, it has rent the solid
rock and in its immediate neighbourhood the violence of the shock
was extreme. Trees have been overthrown or killed as they stood ;
a huge mass of rock, dislodged from near the crest of the hills has
rolled down the slope, scoring the side of the hill. On the opposite
side an equally large block has been dislodged, and in its downward
course cleared a straight track down the hill ; and on the summit a
gap has been cleared by the overthrow of trees along the line of
fracture. It was this gap and the two scores down the hillsides
which appear to have given rise to the report that the hill had been
rent from top to bottom ; at any rate they were pointed out to me
as the course of the fissure.

At first, on seeing what was the true character of the appearances
on which the report was based, the natural inclination was to reject
it as one of the fables which are narrated of earthquake effects; but
the band of trees, killed as they stood, which ran up north-east
side of the hill showed that something unusual had happened, and on
a closer examination it was found that great slabs of weathered
gneiss had been rent in two, and on the crest there was a well marked
depression, like a sma!l ditch, running away to south-west.

This fracture is not, however, confined to this hill, To the north-
east, on the face of the mass of hills to the east of the alluvium, a
line of landslips could be seen, and, on following this up, a zone of
fissuring, accompanied by great overthrow of trees and breaking of
. bamboo clumps, could be traced for about half a mile into the hills
before it became too indistinct to be traceable.

The line of this fracture runs almost under the manager's bungalow,

CHO

150 OLDHAM: GREAT EARTHQUAKE OF 1897.

which was completely levelled to the ground and numerous earth-
fissures opened. Some of them may have been directly connected with
the fracture, as the depth reported to have been measured, 45 feet»
seems greater than is likely to be consistent with an origin in the
manner described in chapter VII.

A feature like thisis naturally more difficult to follow than a
fault scarp, and for some distance to the south-west it is only im:
perfectly marked by a number of small landslips and destruction of
jungle, About 23 miles south-west of the hill there is a forest path

Fig. 16, Sal tree in forest near Bordwar, broken by the earthquake.

leading north and south, along which, on the continuation of the
line of the fracture, there is a well defined band of about half a mile
broad in which cverturned trees are much more abundant than on

( 150 )

THE EPICENTRAL TRACT, 151

either side, and towards the centre of this band the overturned
trees are not only more numerous, but many of the smaller ones, up
to 6 inches in diameter, have been snapped across by the violence
of the shock, as shown in fig. 16.

Still further to the south-west, at the junction of the Lokia Ján
- with the Kulsi, a line of disturbance runs diagonally along and down
the hillside, and, where the soil is shallow, many trees have been
killed, Still further to the south-west a band of overturned trees can
be traced through the submerged forest at Ukiam (see p. 123), but
beyond this I was not able to trace the fracture,

The total length for which this fracture has been traced is con-
sequently about 7 miles, and along the whole of this there is nowhere
any definite and conclusive indication of movement, whether vertical
or horizontal, though at many points there are indications of a small
change of level. The amount is nowhere even so much as one foot,
but where there is any indication at all, the upthrow seems to be on
the south.

Besides the larger fractures, which could be followed for some
distance, a number of smaller ones were seen, in the form of small.
faults of a few inches toa foot or so throw, and lines of fissuring
crossing the path. In many cases it was impossible to determine
whether these were really faults or fractures, or merely incipient land-
slips, and it was only towards the end of the tour, and after the larger
faults had been examined in detail, that I began to recognise those
signs of unusually great, but very local, violence which distinguish
faults and fractures, from the fissures which merely mark an incipient
landslip.

Looking back over my notes and recollections, I think that
many ofthem must have been crossed in the central part of the Garo
Hills, but of those only one was certain enough to be marked on
the map, This is a small fault in the lands of Mandalangiri, in the
Garo Hills, and about 11 mile north of the camping ground. This
fault crossed the path on the crest of a spurand ran in an WNW-

e Co

1:2 OLDHAM: GREAT EARTHQUAKE OF 1867.

ESE direction, transverse to the run of the crest of the ridge, with
a downthrow of about 2 feet to the south. Off the path the throw
could not be recognised, but the lineof the fault could be traced
through a clearing for about 300 yards before it was lost in dense
bamboo jungle. ;

We come now to the consideration of the second class of perma-
nent changes, the lakes or pools formed by interruptions of the
gradient of drainage channels, not directly due to faulting. The prin-
cipal group of these lies in the north-eastern portion of the Garo Hills
District, to the south and south-eastwards of the faults already de-
scribed. In describing them it will be more convenient to start from
the south and take them in the order in which they were actually
seen.

The first of those seen by me was in the valley of the Rong-
tham river, about 3 miles above its confluence withthe Samsang
or western branch of the Sameswari, and in the lands of the village of
Dobukhol. I first saw this at its outlet, and, on being informed that
this was the lake of which I had heard, it was difficult to realise
that there was anything unusual. To the left, or down stream, the
stream bed was ordinary coarse shingle, a natural mountain torrent
bed, to the right stretched a pool which at first sight did not differ
from the pools which are common enough on mountain streams. I
was, however, informed that before the earthquake there had been
no pool here, but a reach of rock and boulders exactly similar to
the portion of the stream bed below the present pool.

Embarking on a bamboo raft on this pool, I proceeded to ex-
plore it. Travelling upstream I found the water gradually deepening,
but the bottom still retained the same character of coarse sub-
rounded boulders, characteristic of astream bed with a rapid current,
‘The deepening was gradual and the presence of a slight deposit of
sand and mud over the boulders pointed toa change in the con-
dition of the stream at this point, from those which had prevailed
when the bed of coarse boulders was originally formed,

E d

THE EPICENTRAL TRACT. 153

As the water in the pool deepened, the lower line of vegetation
on the banks, marking normal high water level, descended to the
water's edge, and then trees were seen standing in the water and
killed by the submergence of their roots, About 200 yards from
the lower end the stream bed takes a bend, and here the greatest
depth of water was found to be 12 feet. On the inner or concave side
of the bend was a gently sloping terrace of stream gravels, which
had been partially submerged, and a number of trees and bamboo
clumps had been killed, their dead stems standing up out of the water
and proving the recency of their submergence. Here the track
from Darangiri crosses the stream and affords independent evi-
dence of submergence, for though there was not ordinarily a greater
depth than a foot before the earthquake this is now the deepest
part of the pool.

About 200 yards above the deepest point, the stream once
more resumes its normal character, and as the water shallows
towards the upper end, the bottom can be seen covered with
boulders. At the head of the pool isa small deltaic deposit of
boulders, due to the checking of the current as it entered the pool.
That there is not a larger delta is due to the other pools, to be
referred to presently, higher up the stream, by which most of its
burden was intercepted.

From the gradual increase in depth at either end it is evident
that this pool is due to a bending of the river bed exactly similar
to that which gave rise to the two large pools on the east side of
the Chedrang fault. In this case no fault was seen, and I do not
think that, had there been one on large enough a scale to account
for this lake, it could have escaped notice. As regards the
amount of the bending that has taken place, the depth of the
pool shows that the stream bed has been depressed at least 12 feet
below the outlet of the pool; there has consequently been at least
that amount of differential movement in a vertical direction, But
the stream Led was not originally level; to judge from its nature,

( age )

1541 OLDHAM: GREAT EARTHQUAKE OF 1897.

above, below, and where it could be seen under, the waters
of the lake, it must have had a fall of about 200 feet per mile.
Taking only one half of this as the actual fall, between the deepest
part of the lake and the outlet, there must have been originally a
difference of 12 feet in level, and as the point which was formerly the
higher by at least 12 feet is now the lower by the same amount, we
reach the conclusion that the total change of levels cannot be less
than 24 feet.

About a mile and a half above this is another pool of similar
character, though not so long or deep, and about three-quarters of a
mile further up is the lower end of another lake, which has a length
cf about one and a half miles and a maximum depth of 18 feet.

This lake or pool is situated in the lands of the village of Lenkra,
where the stream runs through a patch of soft tertiary shales and
sandstones, and the valley is consequently more open and the
gradient lower than at Dobukhol, So fat as can be judged from the
present aspect of the pool, and from the character of the stream bed
above and below it, the stream flowed through an old alluvial plain
of fine grained muddy silt, on a sandy bottom in a narrow, steep
sided channel of some 15 to 18 feet in depth. The gradient must
have been small, but the average must have been increased above
that of the sandy stretches by occasional barriers of rock.

Travelling up this pool from the lower end, the banks remain
high for the first quarter of a mile, and the signs of submergence
are not very noticeable. About three furlongs from the lower end `
there is a depth of 9 feet of water, where a footpath used to cross
the stream and the normal depth was less than a foot. Here dead
bamboo clumps begin to be conspicuous, Half a mile from the lower
end the high banks have sunk beneath the water and the pool
spreads out over the sutface of the alluvial plain, the outer margin
being indefinite, while the channel is bordered on both sides by a
tangle of dead and bleached bamboo stems and tree trunks. This
continues for about half a mile, the margins of the pool then become

( 154 )

THE EPICENTRAL TRACT. 185

more defined and the water gradually shallows to the head of the
pool, where there is an extensive deltaic deposit of sand brought
down by the stream of the Rongtham.

About 3 miles above Lenkra the Rongtham divides into two
branches at the village of Dobu, and in the western of these, just
above the junction, a pool of about half a mile long and a maximum
depth of 10 feet has been formed.

From Dobu I crossed into the drainage area of the Ronghri and at
Náphak found a group of three lakes, in a series of parallel valleys
having a general course of about north and south. Of these the
central and western flow north and unite below the lakes, the
eastern on the other hand drains to the south and joins the central
stream above the lake in it.

The easternmost of the three lakes is in the valley of a small
tributary which apparently carried no water in the dry weather, at
any rate the water was absolutely stagnant, there being no current
over the barrier at the outlet. This valley had an alluvial bottom
which has been flooded for a maximum width of about 100 yards and
a depth of certainly not more than 12 feet, a sounding of ok feet
being obtained at the only place where I could get well out into
the centre. Owing to the character of the valley there is no clear
channel through the lake, but the whole width of the water is occu-
pied by a thicket of dead saplings and bamboos. At the outlet the
alluvium of the valley is very much fissured,

The central lake is larger and more open, the latter fact being
largely due to its having been formed in a hum, or forest clearance
for temporary cultivation. It is about a mile long and about 12 feet
maximum depth. The outlet at the lower end is over crystalline
gneiss, and the water deepens gradually to the lake.

The western lake is rather smaller but of the same type.

The barriers of these two lakes might be ascribed to a single
roll or fold striking about NW and crossing both the valleys
which drain to the northwards. The barriers of the eastern lake
must, however, be attributed to a different fold, for the valley drains

و یچوم

156 OLDHAM: GREAT EARTHQUAKE OF 1897.

to the south and the whole pool lies to the south of the line joining
the barriers of the two other lakes.

With the exception of the two small pools east of Sámin referred
to above (p. 148), these are all the pools, not directly due to
visible faulting, which I saw in the Garo Hills; but there is at
least one other, which appears to be larger than any I saw,
at Sidugiri, on the boundary of the Garo and Khasi Hills Districts.
According to accounts received from Captain Howell and Mr. J. C,
Arbuthnott, C.LE., it has a length of over a mile, a width of 150 to
200 yards, and was too deep to be bottomed with a 20-foot pole,

In spite of careful enquiries made by myself and through the
local authorities I could obtain no intelligence of the existence of
any other pools either in the Garo Hills or in the north-western por-
tion of the Khasi Hills Districts. I do not, however, regard this as
showing that there are no such pools, beyond those mentioned, for I
only heard of two of the four pools in the Rongtham valley, and
of the many ponds formed by the Chedrang and Sámin faults. I only
got intelligence of the one large one at Sámin and the two large
ones in the Chedrang. Ofthese, I only got definite intelligence
when close by while the many smaller pools were not only not
reported, but their existence was positively denied till I found then
myself, and was then quietly informed that they had not been men-
tioned as they were too small to count.

Taking this into consideration, I think that while the existence of
any large lakes or pools in the country off the track followed by me
is unlikely, itis probable that there are many smaller ones.

This conclusion is supported by the fact that, about 50 miles to
the east, I found three small pools of the same type in the northern
Khasi Hills between Rambrai and Nongkhlao.

The first of these is about a mile westwards of the Landomode
Station of the Great Trigonometrical Survey. The water in the Weng-
ing stream is ponded up in a small patch of alluvium, and stood
bank high in the dry weather. Lower down the stream resumed its
normal character, and it was evident that the deep pool was of recent

( 156.)

THE EPICENTRAL TRACT. 157

origin. In reply to an enquiry of my guide I was informed that there
was no pool here before the earthquake.

The next case was at Sinya. East of the village a patch ofallu-
vium, at the bottom of a small valley, which had been taken into
cultivation and made into rice fields, was flooded to a depth of 3 or 4
feet, evidently by an alteration of the general slope of the valley.

The third was at the village of Kanrut, again ina valley with
an alluvial bottom, partly converted into ricefields, Here the plain
had been submerged for about half a mile, and the old ricefields
could be seen, covered by water about 4 feet deep.

It will be seen that the changes here areona much smaller scale
than in the north-eastern Garo Hills, but they are of the same cha-
racter, and the recurrence of the same conditions in this region lends
great support to the supposition that similar features would be ob-
served in the intermediate country, which it was not possible to visit.

Besides the direct and conclusive evidence of permanent defor-
mation which have been detailed, I was able to collect certain evi-
dence, of less value, pointing to changes of level, as inferred from
changes in the appearance of the landscape.

The first of the places where such evidence was noted was at
Mao-phlang on the road from Shillong to Cherrapunji, where
Mr. Evans informed me that, after the earthquake, he noticed a
considerable change in the appearance of the hills to the west. In
part, the statements were general and might be attributed to imagin-
ation or defective memory, but two definite facts can hardly be
explained in this manner.

Beyond Mao-phlang is another mission settlement at Mairang,and on
such occasions as the missionary there visited Mao-phlang a look-out
for the arrival of the party was naturally kept by Mr. and Mrs, Evans.
Before the earthquake, I was informed that only a short stretch of
the road was visible, where it rounded a spur at about 3 miles off;
the crest of an intervening ridge hiding the road before it came out
round the next spur. Now a much longer stretch of the road is

ola ten.)

158 OLDHAM : GREAT EARTHQUAKE OF 1897.

visible, and it can be seen rounding the next spur, where Í was
positively informed the road could not be seen before the earthquake.

The second fact is that a few days after the great earthquake
Mr. Evans took a piece of board and nailed it to a stout post in such
a position that its upper edge was sighted on to the crest of a ridge
about one and a half mile to the west. When I saw it, at the end of
December, six months after the earthquake, the top edge of the board
no longer pointed to the crest of this ridge, but to some way down its
slope. The angle subtended between the point where the edge of
the board then pointed to, and the crest of the hill was about 1°, as
determined by an Abney's level, and the change might be due to a
displacement of the post, though there was no appearance of such.
Apart from this, Mr. Evans informed me that when the board was
put up he could only just see the top of the next ridge, beyond that
on which the board was sighted ; now a considerable stretch of this
can be seen and according to Mr. Evans much more than was visible
soon after the earthquake. !

These facts are of interest as suggesting that no inconsiderable
fraction of the total movements which have taken place, were accom-
paniments of the large number of severe aftershocks. |

The next place where similar evidence was obtained-was on the
road through the Garo Hills to Damra, where it crosses the high
ground north of Cheran and just before descending into the valley of
the Bangshi. Here I was informed by the mouzadar accompanying `
me that beforethe earthquake it was only just possible to see tlie
Brahmaputra over an intervening hill, while now the whole width of
the river was visible. As this statement was a voluntary one, not
made in reply to a leading question, and as a change cf relative level
of the hill and the road of 20 feet atthe outside could be sufficient to
account for the facts, I think it is not improbable that there has been
a change here, It may be noticed that this spot is near the con-
tinuation of the line of the Chedrang fault, and less than 5 miles from
the most southerly point to this which was traced,

(- 15880 J

THE EPICENTRAL TRACT. 159

The third case was at Tura. This has already been referred
to in Captain Howell’s report!, and I took the opportunity, when
passing through Tura, to interrogate some of the military police on
the matter. A system of signalling by heliograph is in use by the
battalion of military police stationed in the Garo Hills, and one of
the stations, with which communication is made direct from Tura, is
Rowmari, on the bank of the Brahmaputra. Before the earthquake
it was just possible to do this from a certain spot by a grazing ray
over an intervening hill. Nowthere is no difficulty at all, and instead
of Rowmari being just visible over the hilltops, a broad stretch of the
plains east of the Brahmaputra is visible, Iwas also pointed out
other hills which were said to have moved up or down since the
earthquake, but attach little importance to those statements. The
statements regarding the changes on the direct line between Tura
and Rowmari deserve more attention as on this line it would be essen-
tial for the signallers to be thoroughly acquainted with the aspect of
the country, and any change would be readily noticed. Having
noticed a change on this line, and so become familiarised with the
idea, it is easy to understand how others, possible purely imaginary,
would be noticed in directions where there had not been the same
necessity for an accurate knowledge of the configuration of the
hills.

Though I made careful inquiries, I could hear of no case of the
drainage of the hills to the west of Tura having been interrupted,
nor did [see any on the road between Tura and Rowmari, except
in the alluvium where they were probably secondary effects of the
earthquake. It may well be that such changes as have taken place
are too small, and extend over too great a horizontal distance, to
have altered the gradients of the streams to an appreciable extent,
and it is to be hoped that the supposed change of levels may be
confirmed or refuted by a retriangulation of this country. Meanwhile

. it may be accepted as probable that there has been a change in the

height of Tura or some of the hills to the west of it.

1 Chapter II, p. 14.
E E :)

160 OLDHAM: GREAT EARTHQUAKE OF 1897.

Reviewing the facts collected, we see that the permanent change
accompanying the earthquake are not confined to one spot or one
line, but extend over the northern part of the Assam hills for a dis-
tance of 100 miles from east to west. Another point to notice is
that, at the limits of this area the evidence points to the changes being
of the nature of long, low rolis, the change of slope being insufficient
to cause any appreciable change in the drainage channels, Then
comes a zone in which the surface changes are more abrupt, the
slopes of the stream beds have been altered so as to cause con-
spicuous changes in the nature of the streams, but any fracture or
faulting which may have taken piace has died out before the
" surface was reached. And north of this, close to the edge of the
hills, the rocks have been fractured and faulted right up to the
surface.

The changes which have taken place are most conspicuous in
the Garo Hills just west of the g1? meridian, and it is here where
the greatest permanent displacement has taken place. The passage
of a zone of folding unaccompanied by faulting into a zone of
faulting and folding is conspicuous, as also the fact that the biggest
of the faults is also the most northern.

On the eastern line the folding, and consequent interruptions
of drainage, is much less, but here too we find the great Bordwar
fracture to the north of this zone—a fault, except in the absence of
noticeable throw.

Seeing then that the permanent disturbance becomes more ac-
centuated to the north, and reaches its maximum at the northern
edge of the hills, we may assume that it does not abruptly cease
along that line, but extends out under the alluvium of the Goalpara
and Kamrup districts. Here, however, we enter a region where
the secondary effects of the earthquake wave, described in Chap-
ter VII, are so extensive as to mask the primary changes of the
nature of those described in this Chapter. For this reason, it is
impossible to get any certain proof of changes of level, other than
those due to shiftings of the superficial layers of the alluvium, but

160 )

THE EPICENTRAL TRACT. det

there are certain facts which point to the conclusion that such have
taken place. |

All down the course of the Brahmaputra the floods of 1897 were
more extensive and rose higher than they had been known to do
before, but nowhere were they so long continued and so disastrous
as in the Borpeta subdivision. Not only were the floods severe
during the wet weather, but even after the dry weather had set in, and
the rivers fallen to their lowest level, large tracts of country, usually
dry, remained under water. It is not to be wondered at that this
unusual flooding was attributed to subsidence of the land, but though,
as will be seen, there seems good reason to suppose that subsidences
and elevations of deep seated origin took place, the greater part of the
unusual flooding must be attributed to the superficial changes de-
scribed in Chapter VII, which were produced by the earthquake.

The regions in which the floods were worst are, it must be noticed,
those where the filling up of river channels, and the shaking down of
the high river banks, was most marked. The effect of these changes
was threefold. First, the subsidence of the high land bordering the
river channels, caused it to sink below flood level and, apart from
anything else, these lands, which are ordinarily left dry, were sub-
merged in 1897. Secondly, the sinking of this barrier along the
river channel allowed its waters to spread more freely over the
low land away from the river. Thirdly, and this was the most im-
portant, the raising of the beds of the drainage channels blocked the
usual means of escape of the greater part of the flood waters, and
necessitated their spreading over the land.

To these three causes the greater part, if not the whole, of the
unusual floods must be attributed. To them is due the fact that the
resthouse at Mankarchar, which stands above ordinary flood level,
was submerged, with a depth of 3 feet of water over the floor, and
that-the floods extended over vast tracts of land, which had not been
submerged before, in Sylhet, to the west of the Garo Hills and in
‘Goalpara and Borpeta.

M | Eër )

162 OLDHAM: GREAT EARTHQUAKE OF 1897.

In investigating the facts and their interpretation it is almost
impossible to separate out from the result of these vastly pre-
ponderating influences, the effects of such permanent changes of
level, as may have taken place. One or two facts, pointing to such
changes having taken place may, however, be noticed.

At Gauhati, a river gauge, fixed on the rocky bank, is regularly
read anda record kept of the height of the river. The peculiar
sudden rise, and subsequent slower fall, of the rivers has already
been noticed’ and for some months, the river levels appeared to
be normal; that is to say, the daily variations in level were too
great to allow any small permanent change to be noticed, In the
months of January and February, however, the water level is
ordinarily very steady and atits lowest, The readings for these
months in 1898, show that the water level was about a foot
higher than usual, and consequently point to a small permanent
change in level, or in the gradient of the river below Gauhati. No
great dependence can, however, be placed on this till the ob-
servations of several years have been accumulated, for the difference
is so small that it might be due to a difference in the season, or
more probably, to the fact that the bottom of the river channel,
which must have been raised through a length of 200 miles, has not
yet been completely scoured out to its old level,

More definite indications of a change of level are to be found at
and below the hill of Hathimora. Here the river banks during the
dry weather were 10 to 15 feet lower than usual, or in other words, the
river bed was that much higher than usual, as compared with the
crest of the banks, In part this is due doubtless to a sinking of the
banks, but not wholly, for a rock which used to stand out of the
water near Hathimora is now covered by the sandy bed of the river,
showing that the river bed has been raised, as compared with the
hill, This banking up of the river bed may be due to a subsidence
of the underlying rocky floor and consequent accumulation of

E i 1 Chapter VIi, p. 107.
Ween

THE EPICENTRAL TRACT. 163

sand, or to a barrier being formed lower down, on whose upstream
side sand has accumulated. 2

Lower down the river there is a stretch of channel which has
caused much trouble in the navigation of the river, the channels
have been ill-defined, constantly shifting, and frequently too shallow
to be passable by steamers of even 3 feet draft.. The stretch pre-
sents all the characters of a region of deltaic deposit, and immediately
below it comes the Kholabandha district, where the floods were
unusually deep and extensive. There is here every appearance of a
depression having been formed, which has not yet been completely
filled up by the river.

From here down to Dhubri there has been no particular difficulty
in the navigation, but below Dhubri there is another stretch of river
where the channels are shallow and shifting.

These facts, for which I am indebted to Mr, R. R. Morgan,
Superintendent of Pilots, point to the conclusion that there have been
unequal changes in the level of the ground, affecting the river
gradient, which are too extensiye to be attributed to superficial
effects of the earthquake on the alluvium, They cannot, however, be
regarded as conclusive, and are discounted by the fact that, though
the river below Hathimora is said to be worse than it has been in
the recollection of the oldest commander of the river steamers,
the same statement is made with regard to the river above Disang
Mukh to Dibrugarh, especially at Sissi Mukh. Here the change can-
not be attributed to any general and deep seated change of levels, but
to the ordinary action of the river, complicated by the effect of the
earthquake on its bed, and it is open to question whether the cause of
the difficulties lower down the river may not be the same.

Whether there has been any general, or local, deep seated change
in the levels of the Brahmaputra valley, north of the Garo and Khasi
Hills must, consequently, remain an open question until such time
' as a reobservation of the stations of the Great Trigonometrical
Survey in the Brahmaputra valley can be undertaken.

Mat ( 163 )

WT
és gx

Lv

164 OLDHAM: GREAT EARTHQUAKE OF 1897.

CHAPTER X.—THE POSITION AND EXTENT OF THE SEISMIC -

FOCUS.

Having completed the description of the data necessary for the
purpose, we may now proceed to a discussion of the nature, situation,
and depth below the surface, of the focus from which the earthquake
started. y

All earthquakes are of the nature of a shock or disturbance com-
municated to, and propagated through, the earth. This disturbance
may originate in many ways: some small earthquakes are believed
to have their origin in the falling in of underground caverns; others
have been attributed to underground explosions of steam ; others
again to volcanic explosions ; but the commonest cause, and that to
which all great earthquakes appear to be due, is a sudden relief
of strain.

According to the generally accepted theory of the constitution
of the earth, it consists of a highly heated central mass, which is
continually losing heat, and consequently contracting, covered by
a comparatively thin surface layer which has already cooled as far
as present conditions permit, and is not capable of further con-
traction to an appreciable degree. Consequently, as the central core
contracts, the outer shell is left partially unsupported, and so thrown
into a state of strain, The relief of this strain may be gradual, or
sudden ; in the latter case it takes the form of a fracture, or a
sudden shift along a preexisting fracture, and is accompanied by

an earthquake, whose extent will depend on the amount of strain |

which has accumulated before yielding took place. Where the
earth's crust has no great power of resistance, the strain will
not accumulate, and earthquakes will be frequent and small.
If, however, the constitution of the earth’s crust, at a part which is
subjected to strain, is such that it does not yield readily, the strain
will accumulate, and, when it becomes too great to be borne any
longer, there will be a sudden and violent relief, accompanied by
fracture and giving rise to a severe earthquake, Whatever may be

(5964 7]

THE SEISMIC FOCUS. 165

thought of the theory of the constitution of the earth outlined above,
it is certain that the actual conditions and processes of nature
are more complicated than indicated, but there also can be no doubt
of the fact that the surface rocks can be, and are, thrown into a state
of strain. Instances are given in almost every text-book, and
it is easy to understand that, if this strain increases beyond the
power of the rock to resist it, there will be a sudden fracture,
giving rise to an earthquake-wave; and also that the longer
the strain accumulates, and the greater the power of resistance,
the greater will be the ultimate shock when yielding does take
place.

Turning now to the consideration of the facts which have been
recorded in the foregoing chapters, the first point to notice is that
this earthquake was not the result of a mere explosion or of the
rending open of a fissure. The evidences of permanent displace-
ment are too numerous, and it may safely be said that, whatever the
nature of the origin of the earthquake may have been, it was in-
timately bound up with a very considerable E of one part of
the earth's crust relative to another.

The next point to be noticed is that these indications of dis.
placement are not confined to one spot or oneline, but are scattered
over a large area. We may consequently regard the earthquake
either as a compound one, having a number of separate foci,
that is to say, as being not one earthquake, but a number of
earthquakes which happened to take place at the same time. Or,
as a more probable alternative, we may regard what appear to be foci
as offshoots from the true focus, which was deep seated, and from
which, what may be called, offshoots ran up to the surface.

There is one form of structural disturbance, consistent with the
_ latter hypothesis, which will afford an explanation of the facts observed,
and that is what is known as an overthrust. An overthiust is in
effect a reversed fault, butediffers in the fact that the fissure along

( 165 )

166 OLDHAM: GREAT EARTHQUAKE OF 1897.

which movement takes place is nearly horizontal instead of nearly
vertical, or steeply inclined. Such thrust-planes, as they are called,
have now been recognised as a common structural feature, and
they seldom, if ever, occur by themselves, but are accompanied by a
number of reversed faults of ordinary type, running upwards, at steep
inclination, to the surface.

In fig. 17 is reproduced a portion of the section east from
Quinaig, as drawn by the Geological Survey of Scotland,’ showing
the thrust-planes T T and the minor thrusts, tt. If this were

Ho

NS INA
LN D «NN SSES é 2
LAN e Le 8 AN E US = d Nis
ORAN y Les Ai AE MOS / Up Mhi 0 A
T£ N : Ss کل = کے‎ 4 Get E M jy 1 li 4 X

Din,

RUM NA 003 2 0000 My i *
Ti T

Fig 17. Section running east from Quinaig, in the northern Highlands of Scotland,

enlarged to the scale of the Garo and Khasi hills, we should have some-
thing very like what has already been described as their structure.

If now there were produced, by any cause, a considerable movement
along the great thrust-plane corresponding to T'—which it must be
recollected is buried deep below, and does not reach, the surface— .
` there would be smaller movements along the secondary thrust-planes
T T and the tertiary ones t t, which might reach the surface as actual
faults, but would very probably die out and give rise to mere changes
of level like those to which the pools in the Garo and Khasi hills
were due; while the movement along the thrust-plane would give
rise to the horizontal displacements observed in the trigonometri-
cally fixed peaks in the Khasi hills,

We have, consequently, an hypothesis which satisfies all the
conditions of the case. It accounts for the permanent change

1 Quart. Jour. Geol. SC . XIV; 418 (1888).

2 See appendix G.
(566. 3

THE SEISMIC FOCUS. 167

being scattered over a large area; it accounts for the faults, the
fractures and the local changes of levels; it accounts, at the same
time, for the horizontal shiftings which have been observed; and it
accounts for the absence of any definite centre of maximum violence
as well as the existence of several local centres of maximum violence
within a general central pleistoseismic area,

At the same time, it introduces no condition which is at variance
with or unsupported by observed facts, and we may safely accept it
as the true theory of the origin of this earthquake.

Though apparently the most probable, this is not the only possible,
hypothesis. The surface features of the Assam range, described in
the last chapter, are compatible with, in some respects they suggest,
the idea that these hills are what the German geologists call
schollengebirge, that is mountains which have originated from
straight up and down thrusts, instead of from lateral compression, like
the Alps and Himalayas. If this is so, the faults by which the fault
scarps are formed would be normal faults, and so far from there
having been any compression, the elevation of these hills would have
been accompanied by an extension of the surface. The state of
strain too, which preceded the earthquake would have been one of
tension and not compression.

The mechanism of the production of this form of mountain is not
properly understood, and a condition of tensile strain in the crust of
the earth would be still more difficult to explain, but the fact of the
existence of such mountains and structure cannot be gainsaid, so the
possibility of the state of tensile strain they imply must be allowed.

If such is the nature of the Assam range, and of the cause of this

! There is no English equivalent of the word ‘scholle’ as used by German geologists. The
German precedent has been followed by some authors in America and England, and the
German words rendered as "block? and * block mountains. The practice seems inconvenient
as it might easily happen trat the same word, block, was required in its ordinary and its special
sense in the course of the same sentence. This would lead to confusion, and I suggest the

word lax, from the Greek word for a block of stone, asa satisfactory equivalent for the word
€ Scholle ’ or * block,’ in its special sense.
( 167 )

168 OLDHAM: GREAT EARTHQUAKE OF 1897.

earthquake, there would be no thrust-plane underlying it, and the
focus of the earthquake would have to be regarded as a complex
one. That is to say, there would be no general focus, but a number
of independent ones, along each fault, and the magnitude of the
earthquake experienced would be due to the simultaneous occurrence
of a number of earthquakes of various degrees of severity.

Whether we regard the focus as a thrust-plane, or as a network
of faults, it practically covered an extensive area. The hypothesis of
a thrust-plane is the simplest to work with, as also the most probable,
and it is that which has been adopted in the following pages.

Having established the nature of the focus of this earthquake,
the next point to determine is the approximate horizontal extent
and boundary of the thrust-plane, or of that portion of it ever which
movement took place. This reservation is important, for it is quite
conceivable that the movement may have taken place along a pre-
existent plane of fracture, whose whole extent is much larger than
that portion over which movement took place. If this is so, it is
obvious that,though we may be able to determine the limits of the latter
area with some approximation to accuracy, there is nothing whatever
to guide us to a knowledge of the whole extent of the old fracture.

It has already been pointed out that the changes which accom-
panied tlie earthquake can be traced right up tothe northern edge of
the Garo hills, and that this alone might lead to belief that similar
changes took place under the alluvium of the Brahmaputra valley. It.
has also been shown that there are independent suggestions, if nothing
more, of changes of level along the course of the Brahmaputra river.?

! There is one hypothesis which might explain many of the observed effects, and that is
the injection of a laccolite or sheet of molten rock. This would account for elevations at the
Surface and for unequal elevations, it might not be inconsistent with horizontal shifting,
and the strain on the overlying crust might in places lead to fracture or faulting. Our
knowledge of the physics of igneous intrusion, small though it be, is, however, opposed to the
idea that such an intrusion could take place with the suddenness necessary to account for the
earthquake, and when we consider the large area and the thinness of the intrusion required
to fit in with the facts observed at the surface, this explanation may be dismissed as physically
im possible.

2 Page 162 ff,

E ps

THE SEISMIC FOCUS. ۱ 169

These indications of a northern extension of the thrust-plane are
further supported by the very large number of secondary earthquakes
recorded at Borpeta and Bijni, which point toan extension of the
focus in that direction. We may, accordingly place the northern
limit to the north of Borpeta with some degree of certainty. Further
east the observations of the Great Trigonometrical Survey show that
Shillong lay within the limit of the area over which permanent
displacement has been observed, which may be taken to practically
coincide with the epicentre, It is not impossible that this may have
extended as far north as Gauhati, which was certainly not far beyond
the limits ; but however this may be, the area over which displacement
has taken place broadens rapidly a short way to the west of Shillong
and, as shown in Chapter IX, Bordwar, at the northern foot of the
hills, is within the epicentral area, Between here and Borpeta we
have no certain indications of its boundary, and to the west of
Borpeta the indications are even more imperfect,

It is evident, however, from the observations in the Garo Hills,
that the zone of maximum permanent change lies just west of the
91? meridian, and it is reasonable to suppose that the zone of
maximum displacement would correspond to the greatest width of
the area of displacement, that is of the epicentre. Assuming, for the
moment, that the displacement was mainly in a north and south
direction, we should expect the northern boundary to trend south
and west from about gi? E. longitude, in a manner corresponding
to its southerly and easterly trend to the east of that line, while, to
judge from the accounts of the effects of the shocks in the Borpeta
subdivision and in the Bhutan hills, it would seem that the northern
limit of the epicentre must have reached nearly to the foot of the
Himlayas, even if it did not run still further north, on this meridian.

Turning now tothe southern boundary. This may be approximate-
ly fixed with some approach to accuracy, 7 we have no direct

; evidence.

In the first place, it is reasonable to suppose that this earthquake.
was directly connected with the tectonic processes to which the

( 169 )

170 OLDHAM: GREAT EARTHQUAKE OF 1897.

origin of the Assam range is due. The elevation of this range has
been shown to be of recent date, and its leading features are so
largely dependent on its structure that the elevation may well be
still in progress. Butthe sharply marked southern boundary shows
that, whatever the process to which their elevation is due, its
geographical limit coincides with the southern edge of the hills, the
northern limit of the alluvial plains of Sylhet. It may be expected,
then, that the thrust-plane has approximately the same limit, and
as pronounced displacement has been detected more than half way
across the hills from north to south, we might, on this ground alone,
draw the southern boundary of the epicentre along the edge of
the hills as far as the Sameswari river and thence westwards along
the southern edge of the high hills.

There is some evidence that this course is the correct one,
for in the Sylhet valley there is no indication of an extension of
the epicentre under the plain. The destructiveness of the earth-
quake was about the same from east to west, as far east as Sylhet,
and there is no place in the southern alluvium at which the after-
shocks were more abundant than at other places east and west.

The marked contrast in these two respects between the northern
and southern valley points to the conclusion that the southern
boundary of the epicentre did not make a sweep to the south,
corresponding to that on the north, but was a nearly straight line,
coinciding or parallel with the southern edge of the hills,

The eastern limit has already been referred to in part, when
describing the northern boundary. East of Shillong the exact limits
cannot be followed with certainty, but the epicentre appears to run
out under the plain to the north of Sylhet. This tract was not visited
by me, but information collected points tosome changes of level. I
was informed that the land on the north bank of the Barhil or Mangai
river, below Dauki (Dowke) which sloped very markedly up to the
foot of the hills before the earthquake, has now apparently a much

1 Chapter 1X, p. 135 ff.
( m» )

THE SEISMIC FOCUS. 171

less slope ; and the low lying portions of a tea garden, about four or
five miles to the south, whichin ordinary years were regularly flooded,
remained dry during 1897. More important is the testimony of
Mr. 1. H. Brownlow of Sandai (Sundye) to the north of Jaintiapur ;
writing on the 3rd August he says: “A very strange thing on the
12th June was the suddenness with which the water lying on
the plain disappeared. Just before the earthquake about one third of
the area within easy vision of the hills was covered with water;
20 minutes afterwards, when I looked, a single small pool was all I
could see,"

The apparent change near Dauki, and the change in the tea
garden south of it, might be, the first imaginary, the second superficial,
and due to displacement of the alluvium by the earthquake. The
drying up of the waters south of Jaintiapur cannot be so explained:
it is the reverse of what happened elsewhere, where the raising of the
beds of river channels and ponds and the formation of vents, caused
water to increase on the surface of the ground, and I have satisfied
myself by enquiries that the drying up observed by Mr, Brownlow
was not a general phenomenon. There is no reason, however, to
doubt the specific statement made, and it is obvious that, if it was
due, as is probable, to a change of level, it must have been local,
for any general change of level would not have led to the water
flowing away from one part towards another.

There is support for the supposition that the epicentre tailed off
in the direction of Jaintiapur in the great violence of the shock at
that place, where it seems from the accounts to have been quite as
bad as at Shillong. At Sylhet, too, the severity of the shock was
very great, but to the eastwards it died off rapidly.

Taking all things into consideration, I consider that the eastern
termination of the epicentre took much the form shown on Map
No. 1. :

Turning now to the western termination, it seems certain that

the epicentre extended into the alluvial plain west of the Garo hills,
but how far it is difficult to say. Considerations of symmetry would

[t . |

LUPIS én, ie
یت‎ LA

172 OLDHAM: GREAT EARTHQUAKE OF 1897,

suggest that the western end had much the same form as the eastern,
and there is a certain support for this in the great destructiveness of
the shock at Rangpur, and along the Kaunia branch of the Eastern
Bengal State Railway. The breaking of the stone fence- -posts
described by Mr. Hayden! is a phenomenon which can only be paral-
lelled in, or close to, the epicentre of this earthquake, and the fact
that the greatest destruction on the main line of the railway was at
and between Nilphamari and Haldibari, all point to there being a
narrow tongue-of epicentre running out westwards, to correspond
with that on the east. This supposition is, besides, supported by
the great number of after shocks felt at Rangpur and Kaunia, and by
the fact that some months after the great shocks small after-shocks
were being felt at these places, which seem to have been of local
origin and not to have been felt elsewhere.

I have now discussed the evidence which can be made use of in
determining the limits of the epicentre, and on map No. will be seen
depicted what I believe to have been its probable approximate form
and extent. It must be understood that this is only approximate, and
that the boundary could not have been so simple and symmetrical in `
reality. There are, however, no means for an accurate delimitation
of the boundary in detail, nor would this, if possible, be likely to
lead to a serious change in the general shape and size.

Taking the dimensions drawn as about the real ones, we find that
the thrust-plane must have had a length of about 200 miles, a maxi-
mum breadth of not less than 50 miles, and an area of between 6,000
and 7,000 square miles.

It is these dimensions alone that give any reason for doubting the
correctness of the conclusions regarding the cause of the earthquake
which has been propounded, as the area far exceeds that of any
known thrust-plane. But in making a comparison it is necessary to
remember that, when a thrust-plane has been laid bare by denudation,
all we can know of it is the line of its outcrop. For a few miles into

1 See appendix A.

( 172 )

THE SEISMIC FOCUS. 173

the earth and a few miles of what has been removed we may be able
to infer, with a greater or less degree of certainty, but apart from
this all we can know is the extent in one direction. Even here the
information may bs incomplete, for denudation may have removed,
or not yet exposed, that part of the thrust-plane where its dimensions
were greatest; or the continuation may be covered up and hidden by
newer deposits,

From this it will be seen that in attempting a comparison we can
only take one dimension, and must leave area out of consideration.
If we do this, it is not necessary to go beyond the great ‘Faille du
Midi’ in Belgium, a thrust-plane which has been traced for 120 miles,
for an instance comparable in size with the thrust-plane inferred as
the cause of this earthquake.

There remains now only the determination, so far as is possible,
of the depth below the surface at which this thrust-plane lay. In
attempting this there is only one method, of the many proposed by
different authors, which can be applied in the present case, and that
is the method proposed by Dr, Aug. Schmidt! based on the observed
rates of travel.

Fig. 18. Schmidt's modification of Seebach's hyperbola.
The principle on which this is based is illustrated by fig. 18,
copied from Dr. Schmidt’s paper. In this the outer circle represents
a section of the earth, and C the focus of an earthquake, the depth

from the surface being much exaggerated to bring out the principle
1 Jahresheft, Ver. f. vaterl, Naturk. in Wurttemberg. XLVI, 227 (1890).

C 05 v)

174 OLDHAM: GREAT EARTHQUAKE OF 1897.

of the method more clearly. The small circle represents a coseismic
line corresponding to the moment when the earthquake first reaches
the surface at the epicentre. Onthe assumption that the earth wave
travels with equal speed in every direction, the earthquake will
have reached every point of this line at the same moment, and the
portions of the radii BA, B'A' which lie outside the circle represent
the intervals of time which elapse between the time when the earth-
quake first reaching the surface at the epicentre, and when it reaches
the points A A’. If now we draw a straight line through the epi-
centre at a tangent to the outer circle, and lay off along it distances
equal to the arcs of the outer circle contained between the epicentre,
and the points where the radii from C cut it, and from each of them
draw a perpendicular line equal to that portion of the corresponding
radius which lies between the two circles, then by joining the ends
of these perpendiculars we get a hodograph, or time curve, of the
earthquake,

Now, it is to be noticed that for each equal interval along the
surface of the earth from the epicentre outwards the time interval
increases, that is to say, the apparent rate of travel decreases until
we reach the point where the radius which runs at right angles to the
seismic vertical cuts the surface. Beyond that, the time intervals
begin to decrease once more, or in other words the apparent rate of
travel increases. As, from the method of construction of the hodo-
graph, the line is more steeply inclined in those parts where the ap-
parent velocity is less, and less steeply where it is greater, the hodo-
graph will be divided into two parts, one concave upwards, the other.
concave downwards, and the point of passage from one curve to the
other corresponds to the place where a line from the focus, at right
angles to the seismic vertical, strikes the surface.

Consequently, if we have the hodograph, the depth of the focus can
be determined by drawing a line through the point of flexure at
right angles to the seismic vertical, when the distance cut off will
represent the depth of the focus from the surface. Lk

In other words, the depth of the focus is the versed sine of the

( 174 )

THE SEISMIC FOCUS. 175

angle whose arc is the distance of the point of flexure of the hodo-
graph from its centre.

As we have seen in chapter IV, the hodograph of this earthquake
shows just those features which it should according to hypothesis»
and the point of change from one curve to the other is situated at
about 280 miles from the centre. A distance of 280 miles on the
surface of the earth corresponds to the arc of anangle of a little over
4°, and the versed sine of 4° amounts to 9'3 miles.

It must, however, be remembered that this result is to be taken
merely for what it is worth, The method, however accurate in
theory, is subject to too many sources of error in practice to be thor-
oughly depended on in the present case, as the time observations are
too few and too imperfect to allow the hodograph to be drawn with
perfect certainty.

A more important source of error may be consequent on the form
of the focus. If we suppose that yielding first took place in the cen-
tre of the thrust-plane and spread outwards from thence, no great
change would result in the hodograph, as the rate of spread of the
fissure would probably be much the same as that of the propagation
of the earthquake disturbance. But if, as is conceivably possible,
there was a preexistant fissure, yielding might have taken place
practically simultaneously at more than one point and spread from
more than one centre over the focal area. In this case the hodo-
graph, as drawn, would no longer give the true distance of the point
of flexure from its centre; an unknown amount would have to be cut
off from the right hand side of the curve and the resulting depth
focus correspondingly diminished.

There is, however, one consideration, not yet referred to, which

° would have an opposite effect and would more than counteract any

diminution in the deduced depth of focus resulting from the supposi-
tion referred to in the last paragraph. In fig. 18 and its explanation it

is assumed that the earthquake travels at the same rate in every di-

rection and at all depths. It is certain, however, that the waves

(125 )

afr d

AP, vw
تو یز‎

176 OLDHAM: GREAT EARTHQUAKE OF 1897.

travel faster at greater than at less depths and, as pointed out by
Dr. A. Schmidt,! of Stuttgart, one consequence of this would be that
the wave paths, instead of being straight lines, would be concave up-
wards. Another consequence would be to throw the point of junc-
tion of the two contrary curves closer tothe centre of the hodograph,
that is to say, the focus would be at a greater depth than the
versed sine of the angle contained between the point of reversal of
the curve and the centre.

If then the curved line on plate XXXIX represents the true
hodograph or time curve, the depth of the focus must be more than
10 miles, But it has already been shown? that the records admit of
another and totally different interpretation, and we have seen that the
dotted line is that which more probably represents the true hodo-
graph, or speed curve, of the earthquake, and is in accordance with
what is probably the true theory of the propagation of the earthquake.

Accepting this theory and the corresponding speed curve as the
true ones, it is evident that the time observations can give no clue to
the depth of the focus, and we must reject the estimate, improbable in
itself, of 10 miles. This, however, leaves us without any definite
indication of the depth, and we have to attack the problem by the

aid of general principles.

Assuming the earth to consist of highly heated matter, and to be
cooling down by a slow conduction of heat outwards, we have it
divided into three portions. First, an outermost skin of rock which
has attained a constant temperature and no longer cools or contracts
any further ; second, a shell of rock which is gradually cooling and
contracting; and third, a central mass which has not yet been affected
by the loss of heat from the surface. Now, the loss of heat by the

second of these zones causes a general reduction of bulk, and the:

rock composing the first or outermost zone is thrown into a state of

compression, The materials of the second zone, however, are con-

tracting in every direction, not in a radial direction only, but also
1 Jahresheft Verein f. vaterl. Naturkunde in Wurttemberg, XLIV, p. 258 (1888).
2 Supra, p. 74.

(O)

THE SEISMIC FOCUS, 177

circumference of each circle, and where this zone passes into the
central mass, which will not yield to its contraction, it must obviously
be thrown into a state of tension, as the tyre of a wheel, which iS
put on hot, is thrown into a state of tension, while it compresses the
framework of the wheel.

From this it appears that while the outermost layers of the earth
must be in a state of compression, there is a certain depth at which
this state of compression is replaced by one of tension, and somewhere
at the passage of the one to the other, there must be a “level of no
` strain," Now it is evident that an earthquake, being the consequence
of a sudden relief of strain, can originate either above or below,
but not at, this level of no strain, and, practically, we may say above
only. At great depths both pressure and temperature increase, and
the effect of either is such as to introduce conditions unfavourable
to a sudden relief of strain. Under great pressure it has been proved
experimentally that solids flow like fluids in the direction of
least pressure, and high temperature introduces a condition of plas-
ticity, even at the surface, which is not conducive to a gradual accu-
mulation of strain and a sudden giving way.

For this reason we may take it that all earthquakes originate
above the level of no strain and, what is more, at some considerable
distance above that level, as compared with its depth below the
surface. It should, therefore, if we had any thoroughly trustworthy
method of determining the depth of the focus, be possible to determine
a minimum value for the depth of the level of no strain.

No method has, however, been proposed which is thoroughly trust-
worthy, and instead of using earthquakes to determine the depth of
this level of no strain, we must use its depth, as independently cal-
culated, to fix a maximum depth for the focus of an earthquake.
According to the calculation of Rev. O. Fisher, the depth of the level
of no strain is not more than about 2 miles from the surface Mr,
C. Davison, however, has pointed out that this estimate is based on

! Physics of the Earth's crust, 2nd edition, 1889, p. 98 : Phil., Mag sth, series, XXXVII,
135 (1894).
Ü ij

EM

178 OLDHAM: GREAT EARTHQUAKE OF 1897.

the assumption that rock expands to an equal degree for an equal
increase of temperature at all temperatures, and that, if allowance is
made for the increase of the coefficient of expansion with an increase
of temperature, this estimate must be raised to nearly 8 miles.!

In both these mathematical investigations a number of assump-
tions have necessarily to be made which either do not, or are not
known to, accurately represent the facts of nature, so that neither can
be accepted as a rigidly accurate estimate. But, after making
allowance for all these sources of error, it is evident that the level of
no strain lies at a depth of only a few miles from the surface; and as
it is highly improbable that in the depths below it the conditions are
such as will allow of the production of an earthquake, we may take it
that all estimates of the depth of focus, which exceed some five orsix
miles must, ¿pso facto, be regarded with suspicion.

In the case of the earthquake at present under consideration the
faults described in Chapter IX show that, in part atleast, the focus
extended right up to the surface, and the presumption is that the
great thrust-plane, which may be looked upon as the principal
focus, does not lie at a very great depth. A similar conclu-
sion might be drawn from the local irregularities of the permanent
displacements of level and position which have been detected over so
large an area, and it is probably an outside estimate if we suppose
the principal focus to have been as much as 5 miles below the
surface,

There is one indication which points to a great depth of origin,
and that is the alleged simultaneous arrival of the aftershocks at
Shillong, Sylhet and Gauhati, as described in Mr. T. D. LaTouche's
report? This statement was so striking that an application was
made to the Telegraph Department to have the observations
continued and extended. They were carried on over two months,
but unfortunately, owing to a failure on the part of the Telegraph
Masters to appreciate the real object of the observations, have only

1 Phil. Mag, 5th series, XLI, 136 (1896).
* Appendix A.

[178-7

THE SEISMIC FOCUS. 179

been of value in the case of Shillong and Gauhati, Details of these
are given in Appendix F, from which it will be seen that the alleged
simultaneity, when critically investigated, means intervals of from
4s. to 1m. of time. It must be noticed, moreover, that the records are
not concordant ; for instance, the shock of 5-48 A.M. (Madras time)
on 25th October is reported by the Shillong office to have been felt
at the same time at Gauhati, while the Gauhati office reports that
it received the signal from Shillong one minute before the shock
reached Gauhati. |

From this itis evident that the time intervals were not always so
short as to necessitate a great depth of focus. They might be ac-
counted for by the focus lying between the two stations or to one side
of the line joining them, in such a way that the distance to be tra-
velled was nearly the same in either case. In those cases where the
shock is reported to have been felt simultaneously at three stations,
it is possible that a large area was set in motion, practically simul-
taneously, along a horizontal thrust-plane.

Taking everything into consideration, there is nothing to neces.
sitate the assumption of a great focal depth, if this is inconsistent
with the conclusion arrived at by other lines of reasoning.

Summing up, then, we may say that the earthquake of 12th June
1897 was caused by a movement along a thrust-plane or thrust-planes
and along secondary thrust-and fault-planes, which had a maximúm
length of about 200 miles and a maximum width of about 50
miles. The depth of the principal fissure cannot be determined, but
was probably not so much as 5 miles below the surface, and numerous
secondary faults and fractures rose from it towards, and in some
cases as far as, the surface of the ground.!

1 This chapter, was written before the results of the reobservation of the triangulation of
the Khasi hills was received. I still consider it represents the most probable explanation of
the earthquake, but the revision of the triangulation has shown that it may possibly be
incorrect. These results were received too late to be incorporated in this chapter, they are
discussed in Appendix G, at the end of which will be found a statement of the extent to which
the conclusions drawn in this chapter are rendered uncertain, though not disproved,

N 2 [ 39. )

0 ۴

180 OLDHAM: GREAT EARTHQUAKE OF 1897.

CHAPTER XI—THE RECORDS OF THE BOMBAY
MAGNETIC OBSERVATORY.

At Bombay, though the earthquake was not felt, it left its
impress, as has already been mentioned (Chapter IV), on the records
of the magnetic instruments in the Observatory at Colaba.

The instruments affected were the declination, horizontal force,
and vertical force magnetographs, and the barograph. Reproductions
of the traces of the first three will be found on plate XL, and the
following description, extracted from the account published by
Mr. N. A. F. Moos,! Director of the Observatory, in his report for
the year 1896, may with advantage be repeated here :—

3. It must be noted here, that the disturbance fortunately occurred just after
16 hours—the time of hourly eye observations; a valuable opportunity, therefore,
was secured, and it became possible to note and study the peculiar behaviour of
the instruments during their disturbances, some of which by the rapidity of their
movements could not have been photographed. There may be some doubt as to
whether the disturbance in the Magnetographs was due to mechanical or magnetic
action, but no such doubt can exist for the Barograph. The disturbance of this
instrument must have been caused by distortional waves, or due to tilting of the
instrument. It will be noticed that the maximum effect of the disturbance, as
photographed in the Barograph trace, followed the largest wave shown by Declina:
tion and Vertical Force Magnetographs by about one minute.

4. With regard to the disturbed trace of the Declination Magnetograph on the
other hand, itis difficult to conceive how condensational or distortional waves could
affect thejsuspended magnet under an exhausted receiver so as to set up large
vibrations in it, simulating magnetic action. The magnet was disturbed mechani-
cally, no doubt, by the seismic waves, and these disturbances were typical of
mechanical action. The three motions (besides the vibratory motion) observed in
the Declinometer at the time of disturbance, showed (1) motion of the whole magnet
parallel to itself (East and West), (2) motion of the magnet aS a whole (North and
South), and (3) the slight bobbing motion of the ends of the magnet up and down,
But neither of these motions would show increase of the scale reading, nor would
vibratory motion, if set up by mechanical causes, continue for some time to show

1 A short note on the disturbance of the magnetical and meteorological instruments at the
Colaba observatory during the earthquake of 12th June 1897. Magnetical and Meteorological
Observations made at the Government Observatory, Bombay. 1896 : 4° Bombay, 1897»
Appendix, pp. (1)—(7).

(430. 3

THE BOMBAY MAGNETIC OBSERVATORY. 181

increase or decrease of declination. It may be noted that the intensity of the
mechanical disturbance was feeble, and the motion observed at the time of the dis-
turbance showed ao evidence of any great, much less violent, movements. Finally
the peculiar character of the trace of this instrument, as will be seen later on, leaves
no reasonable doubt that the disturbance was due to magnetic action, which must
have accompanied the seismic disturbance, possibly as effect of a cause.

5. Thecase of the Horizontal Force Magnetograph is, however, different.
From the peculiar nature of the bifilar suspension, it becomes obvious that the
tension of the wires being a principal factor in the general formula for the con-
dition of equilibrium, it must be affected by any sudden displacement of the points
of suspension. A distortional wave, for instance, suddenly lifting the points of
suspension would result in increasing the tension, and any sudden depression
would for a moment reduce the tension. A lateral movement, also, is likely to
temporarily alter the conditions of equilibrium, and this Magnetograph, therefore,
appears to have been disturbed by the combined action of both mechanical and
magnetical disturbances; in this instance more perhaps by the latter than bythe
former, which was feeble as stated above. The trace shows some evidence of
being affected by two causes superimposed upon each other, specially at the end
of the disturbance, where the blurred trace, probably due to change in the focal
distance caused by a slight displacement of the mirror, appears suddenly to pass
from a faint to a somewhat deep impression, bounded by convex curves.

6. Movement in the Vertical Force Magnetograph now remains to be inquired
into. That the centre of gravity and point of support of the magnet in this instru-
ment, do not coincide, and may cause rotation, and, therefore, movement
of the magnet, is self-evident; but from the conditions cf the magnet and the
quantity of motion observed, it does not seem probable that the motion was
due to mechanical causes. This instrument is under an exhausted receiver
but its knife edge is somewhat faulty, and this Magnetograph, therefore,
is peculiarly sensitive to shocks,—an accidental fall of a lamp-chimney, heavy
tread of visitors, and evena slight knock of the hand results in dislocation of
the curve. And such dislocations being frequent, special care is always taken to
guard the instruments from such accidents. And yet no dislocation of the curve
is noticed during the earthquake, which more or less establishes the fact that move»
ments of the earth must have beén very feeble. But the instrument does show
vibrations, which naturally leads to the inference that they must have been caused
by some magnetic action, the first vibration of which is timed to have taken place
exactly when the Declination trace is just lost by the second wave.

7. And it would thus appear as if the seismic convulsion was in some way the
cause of the magnetic action, the latter phenomenon running parallel to the
former, increasing as it increased and subsiding as it subsided, every seismic wave
having its companion effect in a magnetic wave.

8. With regard to the direction. Since the Barograph would not show any
indication of direction, nor the Magnetographs, all of which appear to have been
disturbed by magnetic action, it is difficult to come to any definite conclusions,
Except, perhaps, in the case of Declination, the suggestive explanations given
later on, if true, would point at least that the direction of the magnetic

E is J)

182 OLDHAM: GREAT EARTHQUAKE OF 1897.

waves, and, therefore, of the seismic waves was inclined more to ** East to West ”
than to “North to South.” From the mechanical disturbance of the magnets of
the two instruments Declination and Horizontal Force, which are suspended at
right angles to each other, the more pronounced motion observed during the dis-
turbance in both, was from East to West. The focal distance of tke latter instrue
ment also appears to be slightly affected, due to motion of the mirror East to West.
It, therefore, strengthens the presumption that the Direction of the waves must
have had a strong Easterly component, and the comparative feeble motion of
both the magnets parallel to North and South direction, shows that the Northerly
component was present, but was feeble. We shall now examine each trace
separately.

9. — Magnetograph.—The photograph shown here is a copy (en-
larged about 23 times) of the trace obtained from the Declination Magnetograph
at the Colaba OGRA The time of vibration of the magnet is 5/33 seconds,
and the original trace allows of an accuracy up to one minute in the determination
oftime. The scale value of the instrument is 28°72 for an inch of tabulation in the
original curve. The curve shows a part of the trace of the 12thJune. The regular
break seen on the trace is due toa metallic fan, which automatically cuts cff light
every two hours for four minutes and a quarter. The middie of this break, there-
fore, represents 16” on the 12th June. The usual sensitiveness of the photographic
paper used at this Observatory has been ascertained by experiments with a steady
light, and found that an exposure of about A to the usual kerosine burners (of about
4 candle power intensity) used, is enough for action; and with regard to impression
of light during the vibration of the magnet, it is found that oscillations of amplitude
of 33 divisions of the scale, equal to a displacement of about 28!minutes of declina-
tion and upwards, are not recorded ; the velocity of the speck of light, together
with the motion of the paper, a little over half an inch in one hour, precludes
photographic action unless the amplitudes are brought within the above limit, and
naturally the first impressions produced are at the extreme ends ofthe amplitude,
where the velocity isa minimum. It is only when the vibrations fall to within
about 13 divisions of scale, that the paper shows the impression of light in the
middle of the curve, that is, where the velocity is a maximum.

10. It is assumed in what follows, that every seismic wave was accompanied
by a temporary disturbance of the magnetic circulation. The first wave, which
appears to have just commenced the disturbance in the instrument must have '
passed Colaba at about 5 minutes past 4. The seismic disturbance then seems
to have grown in intensity, and about 7 minutes past the hour the increased
amplitudes (which by the way, it must be noted, were first performed about a
higher zero showing a decreased easterly declination, movement of the curve in
the downward direction showing anincrease) passed the limit above referred to,
and the trace is entirely lost. Immediately after this, however, the impulsive force
appears to have ceased, and the damper of the magnetograph reducing the
amplitudes (the logarithmic decrement has been ascertained to be about to be
="035), bring them within the limit of photographic action, and the trace is just
photographed at about 9 minutes past sixteen. This marks the time of temporary
luli, for immediately afterwards the trace is lost, to appear faintly again but more

Gest (

THE BOMBAY MAGNETIC OBSERVATORY. ` 183

strongly than before at about 11m. This goes on, but the magnetic action now
becomes less and less intense (presumably therefore the cause of the disturbance
of the seismic waves also) and the amplitude of the vibrations falling within the
limit of photographic action, the record runs beautifully clear after this, and every
wave is registered following its predecessor with more or less rhythmic precision
the intensity of the disturbance getting less and less till, at about 34 minutes past
the hour, the disturbed trace runs into normal curve once more.

11. However, one singular feature in the trace, which is clearly visible in the
latter part of the disturbed curve, must also be explained, vis: the running of
the vibration lines into each other bya zig-zag kind of trace. The most
plausible explanation appears to be the following :—

I2. Assuming the normal curve to be the zero for the time, ordinary vibrationss
caused by momentary disturbance, would take place above and below this zero
with equal amplitudes. But from the nature of the appearance of the zig-zag curve
it seems asif the advancing seismic wave of condensation caused a decrement in the
declination and compelled vibrations about a displaced zero higher up the curve,
while during the receding wave (as it passed away) the reverse and opposite effect
followed, the declination increasing, the vibrations were Caused about a lowered zero
each phenomenon of course having its maximum and minimum effect. Fourteen
such waves inall can be detected on the trace, the 2nd or 3rd of which, more pose
sibly the 2nd, was the most intense, and each complete wave was followed by
another after an average interval of about 2 m. The maximum effect of the
advancing and the retreating parts of the wave being recorded at the average
interval of about 1 m.

Disturbance in Declination Magnetograph.

Maximum effect showing "Maximum effect showing

Wave. decreased declination. increased declination.
و و د اص‎ E
Commencement
of disturbance.
h. m. h. m. h. m.
Ist 16 6 16 7 16 5
2nd 8.5 95
ard 10.5 II'5
4th 12.5 13'5
sth aed 16
6th 17 18
7th 19.5 20'5
8th 215 22'5
oth 24 25
10th 26 27
rith 28 29
12th 30 31 End of disturba
13th 31'5 325 ance.
14th 33'5 16 34

€ i8; )

184 OLDHAM: GREAT EARTHQUAKE OF 1897.

13. It will be seen that the beginning of the seismic disturbance, as recorded
by this instrument, shows that it was somewhat sudden and abrupt, and the rise
tothe maximum after the commencement was within 3 to 4 minutes, while, the
talling off, as shown by the clear trace, appears to havetaken place gradually and
steadily till normal conditions were reached. From the very regular trace it
appears that the time of vibration of the magnet must bear a certain relation to
the period of disturbance, the exact time of vibration being some even multiple of
submultiple of the period, otherwise the regularity of the vibrations would be
disturbed by “ interference. ”’

14. The average displacement of the declination East or West during the 4th
wave, which is well defined, would equal that due toa change of *o0034 C. G. S.
units in terms of force.

15. If by producing the curves on each side we get the probable amplitude of
the most intense wave where the two curves meet, the average displacement
appears to be ‘16 inch, equalto*ooosr C. G.S. units in terms of force, and this
shows that the second wave was the most intense.

16. Zhe Horisontal Force Magnetograph.—The trace hére shows the disturb-
ance in the Horizontal Intensity, which appears to have commenced early, about
23 minutes after sixteen, while the instrument shows that the disturbance did not
end before 36 minutes past sixteen. This must be due to the extreme sensitiveness
of theinstrument. The sudden displacement of the curve at the beginning shows
that the vibrations must have been performed about a displaced zero above the curve,
showing increased intensity equal to about 00004 C. G. S. units, and the action
appears, therefore, more magnetic than mechanical, though both actions seem to
be superimposed upon each other. At 53™ past sixteen the vibrations pass the limit
of photographic action and this perhaps marks the time of commencement of the
more intense oscillations. The trace appears faintly at about 193" past the hour, less
faintly at the extreme ends of the amplitudes than at the middle*of the curve, and
it continues faint till about 26” past the hour, when the trace suddenly darkens
but continues disturbed till 36” past sixteen when the normal curve is resumed.
The time of vibration of the magnet has been ascertained to be 8°, and obviously
as this factor, together with the period of disturbance, must influence the motion
of the magnet, the irregularity in the trace is perhaps to a considerable extent due
to that cause, The scale value of the original trace is 28’ 67" for an inch of
tabulation, and the clear length of suspension of the magnet is about 12 inches.

17. Vertical Force Magnetograph.—The trace shown here is that of the vertical
force magnetograph. As pointed out elsewhere, this instrumentis very sensitive
to shocks resulting in dislocations of the curve. No such dislocation is detected
during the earthquake. The unusual thickness of the vibration trace shows that
the cause of the disturbance could not have been instantaneous, but must have
acted for some considerable time, for vibration of the above amplitude, if caused
by, say, momentary action of a deflector, would die out within? a minute, while
the movements shown in trace lasted each for over three minutes. The first wave
appears to have commenced the disturbance at 6i", attained the maximum at
about 8", and ended about 10" past sixteen. The second commenced immediately
after this, reached is maximum effect at about 12” and ended at 14” past four.

( 384 3

^ THE. BOMBAY MAGNETIC OBSERVATORY. 185

19. The Observatory has two Vertical Force Magnetographs, both of : which
show the two waves clearly. The maximum effect of the first disturbance coincides
in time with the greatest disturbance in the Declination Magnetograph. The
amplitudes of the vibrations above the curve are somewhat smaller than those be-
low the curve (this is more clearly shown in Vertical Force No. 2) which shows
a slight decrease in Vertical Force during the time of disturbance. It may be
noted that the time of vibration of this magnet 1S 5'35 seconds, about the same
as that of Declination magnet.

21. Summa y.— Disturbance of magnetical and meteorological instruments at
Government Observatory, Colaba, Bombay, during the earthquake of 12th June
1897.

Hour. Minute,
. fCommencement , « 16 5
End . e - & 16 7
Most intense, 2nd shock, maximum effect . . e . 76 85
End of Disturbance e e ٠ e SC NS e 6ء‎ 34

Horizontal Force, 1st shock d 2 Lee À ` 2

. . . e > e . a 16 30

Declination, tst shock.

End of Disturbance

3 Commencement NIB 6'5
Vertical Force, 1st shock . " e iis 1 : i em a
2nd shock S e d Commencement . e 16 Io
End . e e . 16 14

Commencement . a- A6 5'5

Barograph, 1st shock ~ : : Maximum effect 2 "D 95

Rug e e a TO 12'5

I-have here reproduced only so much of Mr. Moos’ account as
deals with the observed facts; it will be seen that he attributes the
effect to magnetic causes in the main, and in the original this is
treated at greater length. The matter is one of importance, as it
affects the question of whether the time records obtained from the
disturbance of magnetic instruments by earthquakes may be taken as
accurately representing the passage of the earthquake wave.

The disturbance of magnetic instruments by earthquakes has
been frequently observed, but the cause has been a matter of dis-
cussion, and experiments have been made to determine whether it
was purely mechanical or wholly or in part magnetic. M. Moureaux,
Chief of the Magnetic Observatory of Parc Saint-Maur, Paris, had a
duplicate of his horizontal force magnetograph prepared, substituting

Kee y)

186 OLDHAM: GREAT EARTHQUAKE OF 1897.

a copper bar for the magnet, and found that three earthquakes, which
affected the magnetograph, did not affect the copper bar. But
no certain conclusion can be drawn from this, for, as pointed out by
Prof. Agamennone,' the mechanical conditions of a magnet and
a copper bar, on which the earth's magnetism has no influence
are so different that they could not be expeced to respond in a
similar manner, even to a purely mechanical disturbance in which
magnetism had no part,

Turning to a consideration of the instruments at Bombay, which
are similar to those at any other magnetic observatory, we have three
types devoted to recording, (1) vertical force, (2) horizontal force, (3)
declination.

The vertical force instrument consists of a bar magnet suspended
in a nearly horizontal position on two knife-edges which allow it to
turn in a vertical plane, but not to turn horizontally. This magnet
bears a small mirror, on to which a ray, of light is cast from a lamp
and reflected back on to aslowly travelling band of sensitised paper,
and in this manner a record is kept of the varying inclination of the
magnet due to variations in the vertical magnetic force. As noted
by Mr. Moos, the centre of gravity of the bar does not coincide with
that of the knife-edges ; in the first place it is lower, so as to place the
magnet in a state of stable equilibrium, and in the second place it has
to be a short distance to the south of the knife edges to counteract
the dip and keep the bar approximately horizontal. As a consequence
the effect of the inertia of the bar would tend to cause a disturbance
if it were displaced in either a vertical or horizontal direction ; and
if the disturbance was an alternating one, and the period corresponded -
with that of the natural swing of the magnet, these disturbances,
might accumulate toa considerable amount. The amount of the
displacement of the centre of gravity from the line of suspension is,
however, so small, that this effect may probably be rejected, as has
been done by Mr. Moos in explaining the observed disturbances,

Atti. acc. R. dei Lincei, 4th series, VI, pt. i, p. 21 (1890).
( 196.)

THE BOMBAY MAGNETIC OBSERVATORY. 187

but there is another way in which the disturbance of the trace may
be accounted for on mechanical grounds alone. Ifthe instrument
asa whole is tilted, the lamp and recording paper are raised or
lowered, but the magnet retains its original inclination, and
the position of the spot of light on the sensitive paper is con-
sequently shifted. This is, in fact, exactly what happened, and it
would be possible to explain the appearances produced by a gentle
undulation, or alternate up and down tilting, of the recording
apparatus while the magnet remained steady. The absence of any
interruption in the curve, in the case of an instrument so extraordi-
narily sensitive to the slightest shock as this is described to be, cannot
be held to militate against such an explanation, as it would not be
exposed to anything of the nature of a shock.

The horizontal force instrument consists of a magnet suspended
by two silk fibres, fastened about 4 inch apart from each other on the
magnet, and about 12 inches long. The upper attachment of these
fibres can be, and is, turned round, till the magnet, instead of point-
ing north and south, points east and west; in the case of the Bombay
instrument, the upper pair of suspension points are turned through

about seven-eighths of two right angles. As a consequence of
this mode of suspension the magnet is very sensitive, and a move"

ment, either horizontal or vertical, or a tilting in any direction except
that at right angles to the upper pair of suspension points, will
affect the two fibres unequally and cause the needle to twist more
or less till it resumes a position of equilibrium. Here again we find
that the appearances might be accounted for by purely mechanical
causes, and the earlier commencement and later cessation of the
disturbances might be accounted for by the greater sensitiveness of
the instrument to purely mechanical disturbance from the passing
earthquake-waves.

The declination instrument consists of a simple free swinging bar
magnet suspended by a branchof silk fibres and free to swing in
any direction, Here, as in the vertical force instrument, a slight
displacement of the centre of gravity is necessary to counteract the

[ruo 9

Mk, A KE A wé SFe- 1O

r88 OLDHAM : GREAT EARTHQUAKE OF 1807.

magnetic dip, but the displacement is so small that the centre
of gravity lies practically within the area of the small bundle of
fibres from which the magnet is suspended, and it is difficult
to understand how an angular swing could be imparted to the
magnet by its inertia, while a tilting of the instrument as a
whole would have no effect in producing an angular displacement of
the needle. The only form of mechanical displacement which would
account for the effect of the earthquake on the trace, is a twisting
of the instrument as a whole, or a side-long motion of the recording
paper relative to the magnet. As in the case of the vertical force
instrument when tilted, the magnet would retain its original position,
and the spot of light would reach the paper on one side of
other of the position properly due to that of the magnet.
Apart, however, from such twisting of the instrument as a whole
in a horizontal direction, it is difficult to conceive any explanation,
other than magnetic, to account for the angular displacement of
the needle.

I know of no means of directly testing whether there was any
such horizontal twisting of the instrument. The vertical force
instrument gives no check, for, the motion of the paper being in a
horizontal direction, the only effect of such displacement would be
to cause a slight uncertainty in the time determination, but the
amount of the angular displacement of the declination needle re-
corded --exceeding as it does 28’ of arc—makes it improbable that any
such displacement took place; moreover, though it would be im-
possible to recognise the effect of a small horizontal displacement on
the trace of the vertical force instrument, it is difficult to believe that
so large a horizontal angular displacement as half.a degree of arc
would not introduce anomalies in the trace, which it would be
impossible to overlook.

There is one supposition by which a large disturbance of this
needle might be explained, and that is a succession of small alternating
angular twists, whose period was either identical with, or a smali

(488 >)

ELECTRIC EFFECTS. 189

of aliquot part of the natural period of vibration of the needle.
In this case it is conceivable that they might, if sufficiently prolonged,
in time give rise to a swing of the needle, much in excess of the
actual angular disturbances, but this supposition is improbable,
for it is probable that the wave motion was by no means uniform and
the required ratio between the period of the wave and that of the
magnet would not last for long enough to produce the observed effect.

On the whole, therefore, it is probable that the disturbance of
the magnetic instruments was partly, if not principally, magnetic and
not mechanical; but it is evident, from the time at which they were
felt, that they had no direct connection with the electric disturbances
described in the next chapter. We have already seen (Chap. IV)
that. the magnetic disturbances were felt at Bombay at the
exact time when the earthquake-wave must have passed under
it, and, consequently, if the apparently magnetic disturbance
at Colaba was really, as is probable, magnetic, it must have
originated locally, and as a direct consequence of the passage
of the earthquake-wave. This conclusion is also in agreement .
with the facts recorded in Chapter XV, regarding the general
agreement of time of the disturbance of such magnetic instruments
in Europe as were affected, with that of the passage of the earth-
quake waves, and with the absence of a close agreement of the
times at which the magnetic traces were affected with any of the

phases of the mechanical disturbance.

CHAPTER XII.—ELECTRIC EFFECTS.

Changes in the electrical state of the atmosphere and earth currents,
causing a temporary interruption of telegraphic communication,
have been observed to accompany other earthquakes, but in none,
so far as 1 know, have the latter been so conspicuous and severe
as in the present instance.

Among the reports received from the Telegraph Department
several mention interruptions of communication at about the time

[(. 389: )

190 OLDHAM: GREAT EARTHQUAKE OF 1897,

of the earthquake ; but it is very doubtful whether they were in any
way connected with it, except in so far as they might be due to in-
juries to the wires by falling trees, houses, etc. From Shillong, how-
ever, the reports are of a different nature and the facts sufficiently
striking to justify their being recorded.

Nothing is known of the first great shock, the destruction was
too rapid and complete to allow of any observations being made, but
when communication was being restored, four days after the earth-
quake, it was found that the aftershocks, then still going on, were
accompanied by electric currents of sufficient force to cause
severe shocks to the operators, and sparks on the instruments.

It will be best to reproduce the original reports. The first of
these is from Mr. J. G, Morgan, Assistant Superintendent of
Telegraphs, Shillong Sub-Division, in a letter to the Director General
of Telegraphs, dated 12th July 1898 :—

“ As stated before I noticed no thunder ot lightning on the evening of the 12th.
But when I was at the office, trying to restore communication, in handling the
wires, the Telegraph Master, Signallers and I experienced many electric shocks,
some of them of considerable severity. Nearly all these shocks were sometimes
preceded but more often followed by an earth tremor. Owing to the testing
instruments and indicators being buried under the office ruins, I was unable to find
the direction of these currents. At Nangpoh too, while I was working Shillong on
the evening of the 16th, I noticed much the same thing. The line was being
worked in closed circuit with copper to earth at Shillong, and I observed insulation
several times lasting for periods, varying from a few seconds to a couple of
minutes, in every case an earth tremor made itself feil almost and immediately
after. No thunder or lightning was noticed by me at the time.”

The second is extracted from letter No. 80 of the Telegraph
Master, Shillong, to the Director General of Telegraphs, dated 2oth

July 1897 :-=

“1 should, however, like to note the peculiar phenomenon experienced on the
evening of the first day on which communication was restored, viz., on the 16th
June between 19 and 20 hours. Though working continuously with Gauhati from
14 to about 19 hours (on the closed circuit with batteries at both ends) without
any stoppages, frequent insulations were now noticed varying in duration from 5
to 13 minutes. This continued for about an hour before 1 discovered that in each
case the insulation occurred with a shock of earthquake and that the insulation
continued until the next shock restored the current on the line, lt then appeared

( 190 )

THE EARTHQUAKE SOUND. 191

to me that natural currents of great intensity and alternating polarity were being
generated by each shock of earthquake, which had the effect of neutralising and
restoring the battery current. There being a second line available, it was made
use of as areturn wire, when no further trouble was experienced in working. Since
then earthquake shocks have not had the least influence on our work though
repeatedly felt up to the time of my writing.

In conclusion { would beg to add that between the 12th and 16th June, before
communication was restored, repeated attempts were made by me to get Gauhati
and Sylhet, and it was noticed that severe electrical shocks were felt, often

accompanied by flashes, and in every instance simultaneous with the earthquake
rumble, which immediately preceded each shock."

Leaving out of consideration the explanation proposed, this
account, like the first, shows that there were marked electric currents
accompanying the after shocks in the neighbourhood of Shillong,
which rendered it impossible to utilise the earth as a return for the
current. The intensity of these currents was too great for them
to be explained as a purely mechanical result of the disturbance
of the soil in contact with the earth plates, though this is known to
produce slight electric currents, and the cause must be looked for at

a greater depth and more directly connected with the origin of the
disturbance.

CHAPTER XIIIL.—THE EARTHQUAKE SOUNDS, WITH SOME
REMARKS. ON THE BARISAL GUNS.

Like everything else connected with this earthquake, the sound
phenomena, which are so usual an accompaniment of earthquakes,
were exhibited on an unusual scale. Yet, probably on account of
the greater impression left by the more obvious and awe-inspiring
effects of the earthquake, it has not been possible to collect informa-
tion which will add much to the solution of the many vexed questions
connected with the origin of the earthquake sounds,

Almost the only definite fact to be noted in this connection is
that the earthquake was distinctly heard, though not felt, in the mines
of the Raniganj coalfield. In thissame region the earthquake, as felt
above ground might almost have been called severe ; it was universally
felt and caused some damage to buildings.

) “or 3

۴02 0 9 o
a UY

192 OLDHAM: GREAT EARTHQUAKE OF 1897.

It is well known that the earthquake movement, which can be felt,
is comparatively superficial, and that, away from the epicentre, it
rapidly diminishes in amount, even at small depths. That the sound
was heard underground, though the movements were insensible, seems
to show that the small and rapid vibrations to which it is due, were
not the result, direct or indirect, of the surface undulations, but travel-
led directly through the body of the earth.

Apart from this, the observations are neither sufficiently numerous
nor concordant to lead to any results of value. It has been impos-
sible even to determine, in many places, whether the sound did or
did not precede the movement of the ground ; from one and the same
place I have received accounts stating that the earthquake was
preceded by a loud roar, and others which are positive that no sound
of any kind was to be heard till the earthquake had become severe.

Of the loudness of the sounds in the epicentral tract we have
already had evidence in the account of Mr. J. G. Morgan, Assistant
Superintendent of Telegraphs at Shillong," who found that the crash
of houses, falling within thirty yards of him, was Completely drowned
by the roar of the earthquake, and all the accounts from the epicentre
and its neighbourhood speak of the loudness of the sounds heard at
the time of, and immediately after, the earthquake.

Like the loudness of the sounds the extent of the area over which

they were heard was unusually large. Throughout northern Bengal
and the Gangetic delta many reports mention a rumbling sound as

preceding, accompanying, or following the earthquake. In Calcutta |

the accounts of different observers are contradictory on this point,
some asserting that the earthquake was preceded by a rumbling
noise, others being equally positive that the only noise to be heard
was that due to falling buildings and furniture. As there is good
evidence to show that the rumbling sounds were very generally

heard ruund about Calcutta and to the south in Midnapur, Balasor,

1 See p. 6.

( 192 )

THE EARTHQUAKE SOUN ). 193

and even as far south as Coconada, there is every reason to suppose
that a similar sound could have been heard in Calcutta, and that
those who assert that there was no sound to be heard, either did not
notice it, or confounded it with the crash of falling plaster and brick-
work.

To the west the sounds were distinctly heard throughout the

Santal Parganas, as far west as Gaya and Palamau. At Allahabad,

Jabalpur and Bhartpur rumbling sounds are said to have been
heard, and even as far off as Kotgarh! the earthquake was accom-
panied by a sound like that of the passing of acart. To the east they
are reported from several stations, the most easterly being Bhamo,
though at other stations it is distinctly stated that no sounds were
heard. At Mandalay the reports all state that no sound was heard,
though some also state that the severe secondary shocks which
occurred at about 1 A.M. and 1-30 P.M. of the following day were
accompanied by distinct rumblings. Yothe north it is unknown how
far the sounds were heard, but they were very noticeable in the
Darjiling district, and even at the frontier station of Yatung.

Over the area roughly indicated in the foregoing paragraph,
rumbling sounds appear to have been very generally heard, though

‘there are many places from which negative reports were received.

The reported absence of sound may be merely due to its not having
been noticed, but it may also in some cases represent a real fact, for
it is conceivable that within the area over which the sounds were
generally audible, there may have been places where the local condi-
tions were such that the rapid vibrations were damped, or for some
other reason failed to make themselves sensible as sounds.

So far, mention has only been made of sounds of the ordinary
type, which are variously described as resembling distant thunder,
the passage of a train or cart, or similar sounds; the simile employed
being usually a sound with which the reporter was most familiar.

lIn the case of Kotgarh which is in the Himalayas, it is possible that the sound waves may

have been due to a local sympath.tic shock, started by the passage of the earthquake
waves. : :

0 | (ruga: 7)

194 OLDHAM: GREAT EARTHQUAKE OF 1897.

All accounts agree in selecting some sound of deep pitch, near the
lower range of audibility, of the nature of a continuous rumble or
rattle, prolonged and, as a rule, increasing gradually in loudness and
subsiding as gradually into silence. Besides these sounds, however,
the reports from certain places mention explosive sounds, short and
loud, commencing suddenly like the report of a cannon-shot,

Sounds of this kind have been noticed before, in connection with
earthquakes they have recorded in the case of two previous
earthquakes in India, that of 1869 and of 1881; in the latter case
they were mistaken for signals of distress, and the station steamer
at Port Blair was despatched to search for the supposed wreck.

In the case of this earthquake, the accounts from Kohima mention
that shortly (one account says fifteen minutes) after the passage of
the shock three loud explosive sounds were heard, which seemed
to come from the south or south-west. Similar sounds are said to
have been heard at Wokha and Lumding.

Another account, which may be quoted, is that of Mr. A, E,
English, I. C. S. |

Date, from looking at my diary, it must have been the 12th June: time, be-
tween 5 and 6 p.m. On return from a stroll after game, at jungle camp on bank of
‘ Theingale,’. which is some 7 miles south-west of Kyouko village, and about 19
miles east of Tagaung, I noticed the water in the tank, which was an old river
course containing about 300 yards of shallow water, lapping up against the bank
below my tent. My hunters said it must be elephants bathing, but on looking
there were none at the other end. Some onethen pointed to the trees shaking,
and we knew it must be an earthquake. To the best of my memory, about 3 hour
or less after this I heard one or two distinct booms like cannon shots, and came
out of my tent and joined my hunters and followers who had also come out and
were listening and counting the booms. We counted about 25 distinct booms at in-
tervals of about 3 or 4 seconds, The booms were about as loud and distinct as a
shot from a 7lb. gunturned away from one at, say, four miles or more, if the
country was very flat. j

After these 25 booms there were some less distinct rattling noises.

We speculated on the cause, and self and hunters all concluded it must be
heavy guns firing froma steamer at Tagaung or on the river, At the time no one
thought the sounds were caused by an earthquake.

On asking some cartmen (Maung Bya of Pongon) who arrived next morning
from Pongon (2 miles S. E. of Tagaung) if they had heard the noises and what
they were, they said * Yes, cannon firing at Tigyaing or on the west of the river.’

( 194 )

THE EARTHQUAKE SOUND. 195

At Tagaung I heard that the noises had been Feard from the N. W. It was only
when I made enquiries on board the I. F. Co.’s Steamer Mogaung that, on hearing
there had been no firing of big guns, I concluded the sounds must have been caused
by an earthquake.

At the time of occurrence I thought the firing must have been practice or in
connection with the coming Jubilee. It rained very heavily on 11th, 12th, 13th and

onto 15th.
The shocks were not felt, and except for fthe lapping of the water, would not

have been noticed.
The sounds came to us from westenorth-west of Theingale.
* H * * * H D

The sounds were not at alllikethe rumbling of earthquakes I have heard
before, and quite deceived me and my hunters, who were all especially intelligent

in jungle matters and noises.
» * Li * * * *

Similar sounds are said to have been heard at Sigyaing and
Mawlu in the Katha district, At the former place, four loud reports
were heard before the shock ; at the latter, they were heard afterwards.

Other accounts, from places in Northern Bengal, mention sounds
like the reports of distant artillery, but no reference to them will be
made here,as it has not been possible to make out from the accounts
whether they merely refer to the ordinary rumble, or to the peculiar
sounds now being dealt with.

It will be seen that the short explosive sounds referred to here,
like those accompanying the earthquakes of 1869 and 1881, were
comparatively local in their distribution, the areas over which they
were heard were well separated from the epicentre, and the explosive
sounds, though evidently connected with the earthquake, followed it
after a lapse of many minutes.

The cause of these sounds is at present unexplained, and, though
severalsuggestions might be made, fuller and more detailed informa-
tion is required before it will be possible to express a definite
opinion. Onething, however, seems clear ; that the delay between
the passage of the earthquake shock and the hearing ofthe explo-
sive sounds is too great to be explained by the slower rate of travel
of the sound wave through the air, as in the case of the interval
between a lightning flash and its thunder clap. Were this the

02 C 195 )

سی میس SS‏

196 OLDHAM: GREAT EARTHQUAKE OF 1897.

explanation it would place the origin of the sound at'a distance of
hundreds of miles from the place where it was heard, and, even if it
were possible that the sound waves could travel this distance, it
would necessitate a loudness at the place of origin, and an extent of
country over which they must have been heard, which is much in

excess of what there is any evidence for assuming.

More information is to be gathered from the numerous small
after shocks which were accompanied by, and in many cases consisted
only of, sounds. It may be noticed that these sounds were not, as has
sometimes been stated, confined to the rocky ground but were heard
equally in the hills and on the alluvial plain. 1 have good evidence that
nearly allthe shocks which originate under the Borpeta plain are
accompanied by distinctly audible rumblings, and this even in the
case of those very feeble shocks which are barely perceptible, and be-
ing unrecognised at other stations, appear to be of local origin and
small extent.

During my tour through the epicentral tract I had numerous oppor-
tunities of observing these sounds, which were of frequent occurrence
and in some places almost continuous. Some idea of the frequency of
these sounds may be gathered from a record, kept on 21st, 22nd and
23rd January, of the times at which they were heard at Naphak, in the
Garo hills :—

21st jan. 16-8 l 22nd Jan. 10-47 23rd Jan, 14-15
16-15 11-8 14-28
16-25 II-16 14-39
17-15 13:33 15-8
17-25 13-52% 15-2
18-4 13-57% 15-29
18-9 ove 16-3
18-35 15-54 18-3
19-15 16-25 18-30
20-12 16-35 18-33
21-33 17-6 20-42
21-44* 18-7
22-0 18-12*

22-9 18-33*
22-26 18-55*
22-32 19-51
22-39 20-30
22-41* "c 20733
22-45
22-49

THE EARTHQUAKE SOUND. 197

Thus, in the 23 hours over which the record extends, 48 distinct
rumbles were heard, or at the average of about 50 per diem; and of
these only the 7 marked with an asterisk were accompanied by a
shock that could be felt,

This degree of frequency was much greater than I observed else-
where, and it was noteworthy that at my next camp at Samin, which
was close to the fault described in chapter X, sounds were much
less frequent, not more than 8 or 10 in the day, and mostly accom-
panied by distinct shakes. From Samin I marched northwards to
the neighbourhood of Damra and found the sounds more frequent
again. When working along the great Chedrang fault I again noticed
the fewness of sounds, and more especially of those unaccompanied
by shocks that could be felt, In other words, the sounds, unaccom-
panied by shocks, appeared to be more common where the surface
disturbance had taken the form of bending, than where fractures had
extended to the surface.

This conclusion, and most others regarding the sound phenomena,
must be taken merely for what they are worth. It was impossible to
test them by comparison of simultaneous observations at different
places, and itis possible that the results might have been different
had I been at my different camps on dates other than those
on which I actually was at each particular place.

With regard to the nature of the sounds there was great
variation. Asa rule, they began as a low, almost inaudible rumble,
gradually increasing in loudness, though to a very varying degree,
and then gradually dying out after having lasted anything from 5 to
5o seconds, It cannot be said that there was any connection between
the duration and the loudness of the sounds, some of the most
prolonged never becoming loud, and some of those which lasted
a shorter period being as loud as ordinary thunder at a distance of
two or three miles,

Frequently, instead of commencing gradually, the sound com-

menced abruptly at its maximum loudness and then, after a pro-
: (ez ۸

(7 o 9"
(ës, `

198 ) OLDHAM : GREAT EARTHQUAKE OF 1897.

longed rumble, died slowly off. Occasionally the sound com-

menced abruptly and ended soon, resembling the sound of a small
cannon, or in some cases of a shot-gun loaded with black powder
and fired at a distance of quarter of a mile,

I tried to discover if any variation in the pitch of the note was
observable, and though it did seem in a few cases that the pitch,
as opposed to the volume, was highest at the commencement and
sank as the sound increased in volume, yet no general rule of
this sort could be noticed ; variations in pitch were seldom notice-
able, and when they were, occurred without any regularity of
order, The sounds are in fact so close to the lower limit. of audibility.
that there is little room for variation of pitch, They are indeed
less a note than a rumble, and resemble nothing more than distant
thunder, Of this l had a good proof, for on the morning of the 26th
January when marching from Samin to Cheran, athunderstorm came
up from the south-west, and I found it impossible to distinguish
between the earthquake sounds and the thunder at a distance of one
to three miles.

Very frequently the sound had not the continuous nature of
thunder and was rather a very rapid succession of short sounds, a
rattle or rumble in fact, not unlike the sound made by a cart driven
rapidly over a rough pavement,

An interesting point to determine is whether the sounds travel

through the air, or originate locally. If they travelled any distance

through the air it is obvious that the apparent duration of the sound `

would not correspond with the real duration of the earth vibrations
by which it was caused. The rate of travel of sound through the
air is much slower than through rock and, after the sound vibrations
communicated to the air in the immediate vicinity of the observer
had ceased, he would still hear the sound due to the vibrations
which were communicated to the air ata distance and gradually
reached him by the slower route of travel,

( 198 )

te
,

THE EARTHQUAKE SOUND. 199

Were this the case, we should expect that the sound would
be at its loudest shortly after its commencement and gradually die
away. Toa certain extent thisis the case, but the final dying out
of the sound is usually so short in comparison to its total duration,
that, if explicable in this way, the real duration of thesound waves
in the ground must be within a few seconds of the apparent duration
of the sound.

Another objection to the idea that the prolonged duration of the
sounds is due to its gradual arrival, through the air, from more and
more distant origins, is its lowness, the word being here used with.
regard to the volume, not the pitch of the sound, Quite nine-tenths
of the sounds I heard were never loud enough to travel even half a
mile, about as far as they could travel in 2} seconds, and remain
audible. Hence it is evident that durations of ro or 20 seconds
cannot be explained, as in the case of a prolonged thunderclap, by
the varying distances that the sound has travelled through the air.

As another argument leading to the same conclusion, it may be
said that the sounds appeared to be much more noticeable indoors
than out of doors, ہ0‎ the hypothesis of local origin this is easily
understood, for not only would the walls and ceiling of a room serve
to confine, and so intensify, the sound, but they would themselves
be set in vibration and communicate these vibrations to the air of
the room. It even seemed to me that the sounds were noticeably
more audible in a tent than in the open; here the tent walls could
not communicate sound-waves to the air, but they would serve to con-
centrate those communicated by the ground.

From its very nature this last argument is based on an impression
which is incapable of proof, as itis impossible for the same observer
to be both in and out of doors at thesame moment. The number of
earthquake sounds heard during the for months I was in Assam
was, however, so great that a certain aniount of comparison was pos-
sible; it was moreover a point to which I paid special attention and I
believe that the statement in the last paragraph represents a real fact.

The general conclusion come to as the result of observations

( 199 )

F و‎ OLDHAM: GREAT EARTHQUAKE CF 1897.

''ade by me during the cold weather of 1897-98 was that the
'arthquake sounds very seldom travel through the air, and then only
to a small distance; and, that, though probably heard in the air,
they have travelled through the earth, and are only communicated to

the air in the immediate vicinity of the hearer,
> BI

It would be impossible to close this Chapter without some refer-
ence to those mysterious sounds known as the ‘ Barisal guns,’ which
there seems good reason to suppose are seismic in their origin, and
to which several correspondents have made pointed reference in
connection with the earthquake. A full account of them is, however,
impossible ; the amount that has been published on the subject is so
great and the various hypotheses which have been propounded so
numerous, that a full discussion of the subject would occupy more
time and space than can here be devoted to them.

The earliest published reference to the ‘ Barisal guns’ known
to me is in an account of the Antiquities of Bágerhát by Babu
Gourdas Bysack presented to the Asiatic Society and published in
the Journal of that Society in 1867,? in which they are stated to be
r garded as supernatural salutes fired in honour of a local saint. The
subject came before the Asiatic Society again in 1870, and several
communications will be found in the ‘ Proceedings’ of the Society for
that year. In 2888 Babu Gourdas Bysack presented a further
paper on the Barisal guns to the Society, and as aresult of it a
special committee was appointed by the Society to collect information
and draw up a report on the facts and to discover, if possible, the `
cause of the phenomenon. This report was published in the Pro-
ceedings of the Society for و188‎

! The fact that the sounds generally seem to come from a distance is no real argument
against this conclusion ; it only means that the impression produced is similar to that pro-
duced by loud sounds with which we are familiar, such as thunder, when these originate at a
distance.

2 On the Antiquities of Bágerhát; Jour. As. Soc. Beng. XXXVI, 126-135 (1867).

3 Report on the < Barisal guns ' made at a meeting of the Sub-committee held on the 17th
July 1889 to consider the observations recorded during the year 1888; Proc. As. Soc. Beng.

1889, pp. 199-209, Some further communicatio.'s regarding the Barisal guns will be found

in the Proceedings for 1890,

( 200 )

K

THE BARISAL GUNS. 201

Various are the theories which have been propounded, in the
rumerous communications to the Asiatic Society, in explanation of
the phenomenon. First among them may be placed that which can best
be described as sceptical. According to this the ' Barisal guns” are
nothing butthe sound of firearms, or of pyrotechnic bombs, exploded
on the occasion of marriages or other rejoicings, or of distant thunder,

This explanation may be rejected. In individual instances it may
be impossible to distinguish between the sound of fireworks at a
little distance and the true ‘ Barisal gun,’ yet the universal belief in
the reality of this last shows that we have to deal with a phenomenon
distinct from the familiar sounds by which its explanation is attempted.

Another explanation attributes the ‘ Barisal guns’ to electric dis-
charges or the release of explosive gases under water. This hypo-
thesis is too vague to explain anything, and till some definite
indication is given of the nature of the discharge or-explosion, it
cannot be accepted, especially as there is another and adequate
° explanation.

Of the more definite explanations, which are very probably also
partially true, the most popular appears to have been that which
ascribes the sound to the breaking of surf on the sea shore. This
appears to have been adopted because the‘ Barisal guns,’ in the
Gangetic Delta, seem to come from the south, and the Proceedings
of the Asiatic Society contain more than one explanation of
how a large roller breaking on a flat shore, could produce a sound
which, in certain positions, would resemble that of a distant gun, and
some accounts of how such sounds had actually been heard, the result
of a heavy surf.

This explanation, however satisfactory it may be for places not
too for removed from the coast, cannot be accepted in the case of
the precisely similar sounds heard far inland in northern Bengal.
Here another explanation becomes more applicable, that the sounds
are due to the falling in of river banks. When the Gangetic rivers
are rising in flood the high river banks are largely undermined and

zar)

202 OLDHAM: GREAT EARTHQUAKE OF 1897.

fall into the river with a crash which, according to some accounts,
is at times undistinguishable from the sound of a distant cannon.
This explanation, however, fails, like the last, as an universal
one since the sourds are heard at places and at times when it
is inadmissible.

The subcommittee of the Asiatic Society, saw this difficulty
and attributed the sounds to reflection. That is to say, they
were taken to be sounds originating at a distance so great that
they are inaudible till the sound waves had been reflected from a
concave river bank or hillside and rendered audible by being
brought to a focus.

The acceptance of this hypothesis seems to have been mainly
determined by the fact that the sounds were at times only heard
by one of two neighbouring observers and not by the other. - This
is explicable on the supposition that the one who heard the sound
was in the focus of the reflected sound waves, while the other was
not.! |

The last explanation to be noticed is that which regards the
‘Barisal guns’ as earthquake sounds. We shall return to this, and
all that need be said here is that the peculiar local distribution of the
_ sounds which led the subcommittee of the Asiatic Society to at-
tribute them to reflection, is not inconsistent with their being seismic.
It seems, however, doubtful whether the supposed fact is a real one;
preoccupation, disturbance in the household, distraction from what-
ever cause, or even mere inattention, might lead to the sound not
being noticed by one observer. But even if the sound was inaudible,
and not merely unheard, at one place of observation, this would not
be difficult of explanation.

I have already explained that the earthquake sounds do not seem
to travel any great distance through the air, but to originate locally
in the immediate neighbourhood of the person by whom they are

It may be noted that this explanation was rejected by one member of the subcommittee
on the grounds that the sounds were heard in places where it was inapplicable on account of
the absence of high river kanks to reflect the sounds.

( 20277

THE BARISAL GUNS. 203

heard. Now it isa well established fact that the larger and slower
vibrations which cause an earthquake proper, may vary in amplitude
and period, even at two places only a few hundred yards apart, and
it is reasonable to suppose that the same may be the case with the
more rapid vibrations which make themselves audible as sounds.
The lowest note which can be heard by the human ear corresponds
to a rate of about 32 vibrations a second, anything slower than that
being inaudible ; as the sound of the ‘Barisal gun’, like that of
earthquake sounds in general, is of very deep pitch, and but little
above the lower limit of audibility, it is easy to understand how the
sound might be heard at one place, while at another not far off the
vibrations might be just too slow to be audible.

In all the accounts noticed as yet the phenomenon is treated as
a local one, and though in the later papers it is recognised that the
sounds are not confined to the Bakarganj district, or even to the
neighbourhood of the sea face of the Gangetic delta, yet the expla-
nations offered have all reference to the special conditions of the
Gangetic alluvium. No attempt seems to have been made to utilise
the experience of other parts of the world where similar sounds are
heard..

The ‘ Barisal guns’ are in fact no local phenomenon, but have
been observed in many places, In 1822 the inhabitants of the
island of Meleda, off the Dalmatian coast, were startled by a succes-
sion of explosive sounds, which lasted for some years, and caused so
great alarm that a special commission was appointed by the Austrian
Government to investigate and report on them. The report! is an
interesting and valuable document, not only as a record of the facts
and a careful description of the nature of the sounds—which were
evidently exactly like those of the ‘ Barisal guns’—but also as
containing a full account and examination of all the theories that had
been proposed to account for them. The final result of this

1 Paul Partsch, Bericht úber das Detonaticns Phanomen auf der Insel Meleda bey
Ragusa etc. 8° Wien, 1826.

(205.

204 OLDHAM: GREAT EARTHQUAKE OF 1897.

examination isa demonstration that they were of seismic origin, and
that no other hypothesis would satisfactorily account for the facts.

The ‘guns’ of Meleda appear to have lasted for about five years
and then ceased. It is definitely stated that none had been heard
before 1822, and there is no published account, that I am aware of,
of their recurrence after 1827. It may be that outside these dates
they were merely too infrequent to attract notice, but if the
limitation is correct, they fall into a numerous category of outbreaks
of explosive sounds in other parts of Europe and in America. There
are, however, at least two regions where sounds like the ‘Barisal
guns’ are well known locally and may be regarded, like those of
Bengal, as a permanent phenomenon.

Of the Flemish coast explosive sounds known as mistpeffer
have been studied by Van den Brock, who has published a
number of accounts of them! ; and still more recently Professor
A. Cancani has published an account of the very similar sounds
known as marina in the mountains of Umbria?

It will be impossible here to examine all these accounts in detail,
buta brief notice of the points in which they agree or disagree will
be necessary. In all cases and places there seems to be a general
consensus that they are only heard in calm weather. This probably
only means that when there is a strong wind blowing the sounds are
drowned by those directly due to the wind. Apart from this, there
is no agreement as to the time of year, of day, or of the state of the
weather. The accoun ts from different places differ widely in these
points, and even different accounts from the same place differ radi-
cally from each other.

These discrepancies are not difficult to understand, Except the
short lived observations instituted by the Asiatic Society in 1888, no
attempt seems to have been made to test impressions by actual

1 Ciel et Terre XVI, XVI, passim (1895-96).
2 A Cancani. ٠“ Barisal guns, Mistpaeffers, Marina,” Boll. Soc. Sismol, Ital. 111, 222-234
(1898).

( 204 )

THE BARISAL SUNS, 2085

observations extended over a sufficiently long period of time. Impres-
sions are frequently founded on a very slight basis of fact ; thus the
recurrence of a few ‘guns’ in the early morning, or in foggy weather,
might start an impression that these were the circumstances in which
the sounds were most frequently heard, and, the impression once
started, all cases agreeing with it would, as human nature is consti-
tuted, be more noticed than those which did not agree with it, and so,
unless there was a very great preponderance of the latter the
impression would remain. So too, at a subsequent date, another
person might be struck by the recurrence of ‘guns’ in the late
afternoon, and the impression would arise that they most frequently
occurred at that time.

Taking this into consideration, there does not seem to be any
good evidence that the ‘ guns ’ are specially associated with any time
of day or season of the year, or that they are accompanied by any
particular form of weather, except, as has been noticed, that they are
not generally heard when there is a high wind. Nor is there even
any good reason to suppose that in the case cf each individual dis-
trict, there is any such connection between a special form of weather
and the hearing of the sounds as would imply their being due
to peculiar atmospheric conditions,

There is one point on which all accounts are agreed, that the
sounds are seldom heard singly. They appear to be almost univer-
sally heard in groups of three or more, separated by comparatively
short intervals of time, each group being separated by a comparatively
long interval. This is an important point, for in it the ' guns agree
in mode of occurrence not only with earthquakes which can be felt,
but also with those minute and rapid tremors which cannot be felt
but only make themselves known as earthquake-sounds or sound-
earthquakes.

This brings us back tc what appears to be the only explanation
universally applicable, that is to say, the * Barisal guns ' and similar

( 205 )

206 OLDHAM : GREAT EARTHQUAKE OF 1897.

sounds heard in other parts of the world, are seismic in their
nature,"

Among the many communications I have received regarding the
great earthquake of 1897 there are some facts recorded which point to |
the seismic origin of the ‘ Barisal guns.’ One is that they are reported
to have been very numerous and noticeable about Haldibari and
Jaipur Hat, which are close to the western limit of the epicentral
tract, previous to 12th June 1897. The other, which has come to me
from several independent sources, that they were not noticed in
Barisal during the latter half of 1897, and were certainly much less
numerous than usual,

It is easy to understand that so great a displacement in the
earth’s crust as that which gave rise to the great earthquake of 1897
would lead to a redistribution of strain in all the surrounding regions,
In the deltaic region of eastern Bengal, this redistribution seems to
have taken the form of a relief of the preexisting strains which gave
rise to the small movements that may be regarded as the cause of
the ‘Barisal guns’, and hence, till this strain is reestablished, they
will be less frequent than before,

In ascribing a seismic origin to the * Barisal guns ' it must be dis-
tinctly understood that other causes are not excluded. It may be
that some of the sounds classed as ‘ Barisal guns’ are due to fireworks
or firearms, or to distant thunder. It may be too that in some cases
they are due to surf breaking on a flat coast or to the falling in of
river banks; and it is even possible that they are at times due to
the concentration of sound by reflection, as suggested by the sub='
committee of the Asiatic Society. But behind and beyond all these
causes there is another which is not confined to one spot, nor limited
to the Gangetic delta, but is world-wide and perennial. This can

1 It may be noted that this explanation would account for the impression prevalent in the
deltaic districts, where these sounds are heard, and borne out to some extent by the observa-
tions in 1888 recorded by the snbcommittee of the Asiatic Society; that they are more preva-
lent in the latter part of the rainy season than at other times of the year. At this time the
soil of lower Bengal is completely waterlogged, and in this condition forms a better conducting
medium for sound-waves than the dryer and more porous soil ef other seasons of the year.

( 206: )

ROTATION OF PILLARS, : 207

hardly be other than seismic, using the word in its widest sense ; that
is to say, the sounds are due to disturbances in the interior of the
earth, caused by readjustments of the distribution of the material
of its crust, and taking the form of elastic vibrations, which are trans-
mitted to the surface and there become sensible as sounds.

CHAPTER XIV.—THE ROTATION OF PILLARS AND
MONUMENTS.

That pillars and other similar objects may be left standing, but
with one part twisted round upon another, has long been known
as a fantastic effect of severe earthquakes, and even in some cases
of earthquakes which can hardly be called severe. There has,
however, been no instance where cases of this kind have been so
numerous, and so various in the nature of the objects rotated, as
the earthquake now under consideration, and advantage has been
taken of this to collect all the information possible, and to apply
the data so obtained in testing the sufficiency of the various theories
which have been advanced by different authors to account for the
phenomenon,

In treating this branch of the investigation of the earthquake,
I shall first of all describe those instances of rotation which have
come under my observation, and then proceed to an examination of
the theories which have been advanced.

The description, however, requires a few words of introduction,
In order to determine exactly what has happened we require to
know (1) the amount of the twisting, (2) its direction, and (3) the
amount and direction of the lateral displacement of the upper por-
tion as a whole, that is to say the distance and direction that its
centre of gravity now lies with regard to its position before the earth-
quake. The first and third of these require no explanation, but the
second is one in which confusion is possible ; in the case for instance
of a square pillar whose sides now stand at an angle of 40? with their
original direction it is obviously important to know whether it has

(209 3

208 OLDHAM: GREAT EARTHQUAKE OF 1897.

been turned 50° in one direction or 40° in other, This cannot
always be made out from the plans, but can always be seen on the
spot, as it is easy enough to determine which side of the upper por-
tion originally coincided with which side of the lower. The direction
of rotation was, consequently, determined in every case by observa-
tion and is stated in the following description as positive or nega-
tive; by positive is meant rotation with the sun, or in the same
direction as the hands of a watch lying face upwards on the ground,
by negative is meant rotation in the opposite direction.

The most imposing and striking of the numerous instances of
twisting is that of the monument to George Inglis, erected 1850, at
Chhatak. This conspicuous landmark takes the form of an obelisk
and rising from a base 12 ft. square, must have been over 60 ft. high
before the earthquake. It is built of broad, flat bricks, or tiles, laid in
mortar and plastered over, and is represented, in its present state,
in the frontispiece, while a plan and elevation will be found on Plate
XXXV. |

Owing to the size it was impossible to get exact measurements.
The height was determined by means of an Abney's level, and the
position of the twisted portion had to be determined by perpendi-
culars dropped from its corners on to the pediment. With these
qualifications the following particulars may be accepted as substan-
tively accurate.

The topmost 6 ft, 2in. was broken off and fell to the south, while
the next 9 ft. was thrown to the east, as shown in plan on
Plate XXXV. Of the remainder, the top 22 ft. has been separated
at a height of about 23 ft. from the ground, and twisted negatively
through 30°. The centre of the upper portion lies 12 ft. to
N 55? W of the lower, and the section at the fracture is و‎ ft. square.

The upper part has evidently rocked on the lower, as there is
splintering at the corners and edges, and below the fracture a slice
of masonry about 15 in. thick, which was not bonded into the main

( 208 )

ROTATION OF PILLARS. 209

mass, has separated on account ofthe pressure on its upper end.
This separation is on the east side, and is shown in the frontispiece.

At Cherrapunji there are more instances of twisted objects. At
‘Inglis? bungalow a square masonry pillar, originally intended to carry
an aqueduct across a small water-course, 7 ft. و‎ in. high by 2 ft. square
in section, oriented N 30° E, has twisted negatively 5%. The centre
has shifted 3 in. to N 55° E,

Close to this is a pillar rising from the top of the compound wall.
The wall is 3 ft.6 in. high, and the pillar rises 5 ft. above it, with a
section 2 ft, 2 in. square, oriented N 58? E ; this has twisted positively
4°. The centre has shifted ri in. to S 40? E. Plans of these two
pillars are given on Plate XXXV.

David Scott's monument has about 6 ft., of the part left standing,
broken off and twisted about 1? negatively. The corners of the stones
at the plane of twisting have been flaked off owing to the pressure
brought to bear upon them.

At Shillong the gate pillar of the Ferndale Hotel (Plate XXXIV)
was twisted negatively. The pillar was oriented originally N 8? E.
It has turned negatively on its base 4°, and again above the lower-
most stone through another 2°. The centre has shifted slightly to
N. Wi; but as each block has an iron pin fixed in its upper
surface, which fits loosely into a hole in the base of the block next
above it, no great lateral shifting was possible,

The gate pillars of Beauchamp Lodge, represented in the lower
half of Plate XXXIV, have both been twisted. They were 8 ft. high
by 2 ft. gin. square, oriented N 15? E. One pillar has twisted posi-
tively 3°; the centre has shifted 4 in. to N 36° W. The other pillar
has twisted positively 4°, the centre having shifted 11 in. to N 70° W.,

At Gauhati are the largest number of twisted objects. Views
of some of these are given in Plate XXI, elevations on Plate XXXVI,
and plans at the plane of twisting on Plate XX XVII.

Beginning with the group of tombs represented on Plates XXI
and XXXVI we have—
First, an obelisk on the left, No. 3 broken at its base and twisted

E ( 209 )

"UP Ay CPUS, 9 ٦

210 OLDHAM: GREAT EARTHQUAKE OF 1897.

negatively 2°; centre shifted 3$ in, to W 20? S ; section at fracture
2 ft. square.

Second, the tomb No. 2 of the same type, but broken higher up
and twisted positively 48? ; the centre has shifted 2} in. to
N 80? W ; section at fracture 1 ft. 6 in. square. |

Third, the cylindrical fluted column No. 1. The twisting of this can
be recognised by the fluting, but more distinctly by the present
oblique position of the name tablet, which was doubtless originally
placed in the same line with the rest of the tombs. As the undis-
turbed tombs close by only vary in orientation of the corresponding
sides between W 15? S and W 18? S, while the tablet on the column
bears W 24? S, it is evident that this has been t wisted zegatzvely at
least 7°; the centre has shifted $ in.to N 20? E ; the section at
the fracture is 2 ft, 1 in. diameter.

Besides these, the tomb of Samuel Fleming, of the same pattern
as those marked 1 and 2 above, has been broken across the pediment,
and twisted negatively 4°; the centre has shifted 1 in. to N 20° W.

We may now turn to the gate pillars of the telegraph signallers'
quarters, one pair of which are represented in Plate XXI, fig. 2. These
are all square pillars 2 ft. 3 in. across at the plane of fracture, and

about 7 ft. high, oriented N 20° W. In the case of the eastern gateway

represented in Plate XXI, the plane of fracture was 1 ft. 3 in. from the `

sround; in the case of the single pillar of the western gateway left
` standing, it was 2 ft. 6 in. from the ground. They have all been
twisted negatively.

The E pillar, E gate, through 5°; centre shifted 1 in. to S 20° E.

The W pillar, E gate, through 12°; centre shifted 1 "in. to
W 25? S.

The W pillar, W gate, through 6°; centre shifted 21 in. to N 25? W.

At the entrance to the compound of the Commissioner's house
were two large square pillars 10 ft, high by 2 ft, 5 in. square, oriented
N 20° W. The northern pillar has twisted negatively 6°, and the
centre has shifted $ in. to W 20° S.

( 210 ) |

ROTATION OF PILLARS. 211

Besides these, there must be mentioned the pillar in the wall of
the compound W of the Deputy Commissioner's bungalow}, which was
oriented N 25° W, and twisted positively 10°.

At Tezpur, a small iron safe, standing on a four-legged stool in the
corner of a room in Surgeon-Major Macnamara’s house, was turned
negatively through about 40°; a despatch box resting on similar
stool in .the opposite corner of the room was also twisted negatively
through nearly 90°. In both cases the stool turned with the safe
or box resting on it, and, so far as could be judged, almost on
its axis. A newly built porch of this house was broken down
and the brick pillars supporting it twisted negatively, also nearly
on their axis, so far as I could judge from the description.

At Darjiling we have the tomb described by Mr. Hayden (Appendix
A) and represented on Plate XXXII, fig. 1, the plan at the plane of
twisting being shown in fig. 31. The orientation of this was E 40° N;
it was twisted positively 2°, and the centre shifted 4 in to E 40° N.
The section at the plane of twisting was 1 ft. 64 in. square.

At Calcutta some of the pinnacles of the High Court, of octagonal
section, are reported to have been broken across, and the upper part
twisted round. Unfortunately no measurements were made before
these were repaired, and it is not possible to say in what direction or
to what extent they were displaced.

In addition tothese instances of the rotation of untramelled pillars,
the case of the thermometer- shed at Dhubri may be mentioned
(see fig. 24, Appendix A), where aset of brick pillars supporting a
thatch roof were broken off, and rotated posztively from 9? to 12%, At
the Borjuli tea garden, in the Tezpur district, a leaf house whose roof
was supported on brick pillars was affected in the same way; the
pillars were all broken off and twisted positively from 6? to 10”. I
was informed that the same thing had been noticed on several

' See Mr. LaTouche’s report, Appendiz A: the monuments and pillars at Cherrapunji,
Shillong and Gauhati, referred to in the text, were mentioned in his report and afterwards re-
examined by me and measured witha special view to determining the amount and direction
of the displacement of the centre of gravity of the rotated portion.

Leo عو‎

212 OLDHAM: GREAT EARTHQUAKE OF 1897.

other gardens, even as far as the Sibsagar district, but have no

further particulars.

The earliest instance of the rotation of objects by an earthquake,
of which F can find any published record, is in the account of the:
Calabrian Earthquake of 1783 by Michele Sarconi, Secretary to the
Neapolitan Academy of Science and Letters? His figures of the
dislocation of the obelisk-shaped pillars at either end of the facade of
the Certosa di S. Bruno have been made classic by their reproduction
in Sir C. Lyell's “ Principles of Geology," and in the description
is stated, what appears to have been overlooked by most of his
successors who have proposed or accepted other explanations, that
one of the obelisk-shaped pillars was twisted from right to left, and
the other from left to right; while the twisting is accepted as evidence
of the “arrizontale vorticoso” nature of the shock, This explanation;
the only one prevalent at the time, remained unquestioned, tiil
the publication of Charles Darwin's “ Naturalist's Voyage round the
World," where (Chapter XIV), after noticing the effect of the earth-
quake of 20th February, 1835, in causing the rotation cf some square
objects onthe coping of the Cathedral walls at Concepcion he says :—
* 'This twisting displacement at first appears to indicate a vorticose
movement beneath each point thus affected ; but this is highly im-
probable." He suggests as an explanation that it was caused by a,
tendency in each stone to arrange itself in some particular position
with respect to the lines of vibration in a manner somewhat similar to.
pins on a sheet of paper when shaken.

The next attempt to explain this phenomenon is that of Mr. R.
Mallet,? who, rejecting the vorticose theory and the suggestion of

Mr. Darwin, proposed an explanation of his own which has found its
way into most text-books, and may be given in his own words.

* If a stone, whether symmetrical or otherwise, rest upon a given base, and
motion be suddenly communicated horizontally tothat base in any direction, the

1 * Istaria de Fenomeni del Tremoto avvenuto nelle Calabria, e nel Valdemone,.nell ’ anno

1783 ; posta in luce dalla Reale Accademià delle Scienze, e delle Belle Lettere, di Napoli.
Folio, Naples 1784, p. 62 and Pl, XXI.

2 Jour : Geol. Soc. Dublin III, 138-144 (1845). See also his paper on the Dynamics of
Earthquakes : Trans, Roy, Irish Academy X Xl, 51-105 (1846).

( 212: )

ROTATION OF PILLARS. 213

stone itself will be solicited to move inthe same direction. The measure of force
with which the movement of the base is capable of affecting the stone or other ins
cumbent body is equal to the amount of friction of the latter upon its base, a func-
tion of its weight, which, without the intervention of cement, may be from one-
fifth to one-tenth of the weight of the body, for cut stone resting on cut stone, but
may be increased to any amount by the intervention of cement. The stone, how-
ever, is possessed of weight, and therefore of inertia, that is to say, being at rest, its
whole mass cannot be instantly brought into motion by the plane on which it
rests ; and if the amount of adhesion between the stone and its bed be less than the
inertia due to any given velocity of horizontal movement of the bed, the bed will
move more or less from under the stone, or the stone will appear to move ina cons
trary direction to that of the motion of its bed.

Now the inertia of the stone, which is here the resisting force, may be considered
to act at its centre of gravity. The impelling force is the grasp of the stone which
its bed holds of it by friction or adhesion, and this may also be referred to some one
point in the surfaces of contact, which we may call the centre of adherence.

If, then, a stone, or other solid, rest upon a plane which is suddenly moved
with sufficient velocity to produce motion in the incumbent body, three several
conditions of motion of the body may occur, according to the respective positions
of its centre of gravity and of the centre of adherence :—

(1) The centre of gravity of the body may be at sucha height above the base
that it shall upset by itsown inertia. This is the case with houses, towers, walls,
etc., etc., when they fall by earthquakes, accompanied, however, by dislocation of
their parts.

(2) The centre of adherence may be ina point of the base plumb under the
centre of gravity of the body, or in a vertical plane passing through its centre of `
gravity, and in the direction of motion of the base.

In thiscase the stone will appear to movein the opposite direction to that in
which the base has moved, that is to say, the body may have acquired more or
less the direction of motion of the base, according asthe motion of the latter has
been of longer or shortér continuance, or less or more rapid ; butin so far as the
movement in opposite directions has taken placé, the base in reality has slipped
from under the body.

(3) The centre of adherence may neither be plumb under the centre of gravity
of the body, nor in the plane of motion passing through its centre of gravity, but
in some point of the base outside the line of its intersection by the plane ; in which
case the effect of the rectilinear motion in the plane of the base will beto twist
the body round upon its bed, or to move it laterally, and twistit atthe same time,
thus converting the rectilinear into a curvilinear motion in space. The relative
amount of the two compounded motions being dependent upon the velocity and
time of movement of the base, and upon the perpendicular distance measured
horizontall y atthe surface of adherence between the centre of adherence and the

, centre of gravity of the body.

This latter case is that which appears to have wee the „stones of Calabria,
South America and Greece, and affords, as 1 feel assured, the true explanation of
the phenomenon."

(213)

VAR Ai Va 7 7
yee O ZEN

214 OLDHAM: GREAT EARTHQUAKE OF 1897.

Subsequently, when investigating the Neapolitan earthquake of
1857, he found that at the Certosa de St. Lorenzo, near Padula, a
number of pinnacles had been twisted all in the same direction, and
recognising the inapplicability of his former hypothesis, proposed an
alternative one, as follows 1 : —

* Now there are two distinct trains of earthquake causation, by either of which
bodies may be twisted on their bases. First, by the action of a single shock, when
the centre of adherence of the base of the object, lies to one side or other of the
vertical plane passing through the centre of gravity, and the line of the wave-
path. Second, by the conjoint action of two closely successive shocks. By the
first shock, the body is tilted up from its base, but not overthrown, so that for.a
time, greater or less, it rests wholly upon one edge of its base ; while thus poised
if another shock bear upon it, in any direction transverse tothe first, it acts as
usual, at the centre of gravity of the body, to displace it by inertia, in the contrary
direction to the wave transit ; but the body is held, more or less, by friction a£ the
edge momentarily in contact with its support, and there only ; but this edge must

Fig, 19. Diagram (after Mallet) to illustrate the effect of two shocks in causing rotation.

always lie to one side of the vertical plane passing through the centre of gravity
in the direction of the wave-path : hence the tilted body, while relapsing upon its
base, also votates, round some point situated in the edge of its base upon which it
had been tilted, and thus it comes to rest in a new position, having twisted more
or less round a vertical axis.

1 R, Mallet, The great Neapolitan earthquake of 1857. London 1862, Vol. T p. 376. The
same explanation is given by R. Falb, apparently independently arrived at, in his Gedanken
und Studien über den Vulcanismus, 8°, Graz, 1875, p. 223.

( 214)

ROTATION OF PILLARS. 215

if the observer look due south at a square pyramid, for example, whose sides
stood cardinal, and it be tilted bythe first semiphase of a shock from east to
west, the pyramid will tilt or rise upon the eastern edge of its base; and if, before
it has had time to fall back, it be acted on by another shock from north to south,
the pyramid will rotate, upon the bisection or on some other point, of the edge
on which it momentarily rested, and will hence come to repose, after having twisted
from left to right, or with the hands of a watch.

If the tilting up, had been produced by the second semiphase, of the same shock
from east to west, then the pyramid would have risen upon the western edge of its
base, and the same direction (north to south) of second shock, would have produced
rotation upon that edge, but in a contrary direction to the proceeding or from right
to left or agaizst the hands of a watch.

Again, if, on the first supposition, the first semiphase of the east to west shock
had tilted the pyramid upon its eastern edge of base, but the second shock had
been from south to north, in place of the reverse as before, then the rotation would
have been from right to left; and if tilted by the second semiphase on the western
edge, the second shock, south to north, would produce rotation left to right.

It would, therefore, appear at first impossible, to determine the direction of
motion in transit, of either shock, from such an observation : we can, however,
generally discoverupon which edge ofthe base any heavy body of stone or
masonry has tilted, by the abrasion or splintering of the arris, and the rotation
must have taken place round some point in tbat edge. If, therefore, we know
the direction of either one of the two shocks, we can always discover that of the
other, by the rotation observed, and if the time of oscillation of the body be ascer-
tainable, we are enabled to calculate a major limit, for the interval oftime that
must have elapsed, between the arrival at the twisted body of the first and of the
second shock, when both the wave-paths are known.

With a single instance of such twisting, it may be impossible to decide
whether the twist has been due to one shock (1st case),or to two shocks in suc-
cession (2nd case) ; but when several bodies alike or dissimilar, at the same
locality, are all found twisted in one direction, itis certain to have been the work
of two distinct shocks, for it is beyond the reach of probability that several
bodies should al? happen to have their respective centres of adherence, at the
same side of their respective centres of gravity, and unless they have, some will
rotate in one, some in the other direction by any single shock ; rotation thus pro-
duced, being always by the centre of gravity moving contrary to the first or
second semiphase of the wave and carried round the centre of adherence, by the
line joining them as a radius vector ; the inertia of motion at the centre of gravity,
and the resistance of the point of rotation in the edge of the base, or of the centre
of adherence, forming in every case, the extremities of the dynamic couple, ”

There remains one more explanation to be mentioned which, on
account of its simplicity, has been very generally adopted. It is that
of Mr. Gray and contained in Professor J. Milne’s account of the
Japanese Earthquake of 22nd February, 1880. After mentioning a

( 215 )

AA "ZS

~

216 OLDHAM: GREAT EARTHQUAKE OF 1897.

number of instances of tombs that had been twisted, he goes on to
say l :—

“Although fully recognizing the sufficiency of two transverse shocks to produce
the effects which have been observed in Yokohama, I will offer an explanation of
this phenomenon which was first suggested tome by my colleague Mr Gray,
which appears to be simpler than any with which I am acquainted.

If any columnar-like object, for example a prism of which the basal section is
represented by A, B, C, D (see fig.20) receives a shock at right angles to B C,
there will be a tendency for the inertia of the body to cause it to overturn on the
edge B C. Ifthe shock were at right angles to D C, the tendency would be to
overturn on the edge D C. If the shock were in the direction of the diagonal C A,
the tendency would be to overturn on the point C. Let us, however, now suppose
the impulse to be in some direction E G, where G is the centre of gravity of the body.
For simplicity we may imagine the overtürning effect to be an impulse given through
G in an opposite direction, that is, in the direction G E. This force will tend to tip
or make the body bear heavily on C, and at the same time to whirl, round C as an
axis, the direction of turn being in the direction of the hands of a watch. If, how-
ever, the direction of impulse had been E—G, then although the turning would still

have been round C, the direction would have been opposite to that of the hands of `

a watch.

Fig. 2o. Diagram to illustrate Gray's theory of rotation.

To put these statements in another form imagine G E, to be resolved into two
components, one of them along G C and the other at right angles G F. Here the
component of the direction G C tends to make the body tip on C, whilst the
other component along G F causes revolution.

Similarly 6 E may be resolved into itstwo components G C and GF, the
latter being the one tending to cause revolution.

From this we see that if a body has a rectangular section so long as it is acted
upon by 2 shock which is parallel to its sides or to its diagonals, there ought not to

1 J. Milne. The Earthquake in Japan of February 22nd, 1880. Trans. Seismol. Soc., Japan,
L, Ft. Il, pp. 33 35 (1880).

( 216 )

ROTATION OF PILLARS. 2

be any revolution, . If we divide our section A, B, C, D up into eight divisions by
lines running through these directions, we shall see that any shock the direction
of which passes through any of the octants which are shaded, will cause a positive
revolution in the body; that is to say, a revolution corresponding in its direction
to that’of the movements of the hands of a watch; whilst if its direction passes
through any of the remaining octants, the revolutions will be megative or opposite
to that of the hands of a watch.

From the direction in which any given stone has turned, we can, therefore, givé
two sets of limits between one of which the shock must have come.

Further, it will be observed that the tendency of the turning is to bring a
stone, like the one we are discussing, broadside on to the shock; therefore, if a
stone with a rectangular cross section has turned sufficiently the direction of a
shock will be parallel to one of its faces, but if it has not turned sufficiently it will
be more nearly parallel to its faces in their new position than it was to its faces
when in their original position,

If a stone receives a shock nearly parallel with itsdiagonal on account of its
unstability, it may turn either positively or negatively.

Similarly, if a stone receives a shock parallel to one of its faces, the twisting may
be either positive or negative, but the probability is that it would only turn slightly,
whereas in the former case, where the shock was nearly parallel to a diagonal,
the turning would probably be great. ”

We have then four alternative hypothesis, for the suggestion of
Mr. Darwin is too vague and indefinite to rank as an hypothesis or
explanation. These are:—

(1) The theory which is implied in the term vorticose, and
attributes the rotation toa rotary or circular motion
cf the ground during the earthquake.

(2) Mallet's theory, which attributes the rotation to the centre
of resistance and the centre of gravity not being in the .
same vertical plane with “the direction of the wave
motion, and to the stone being consequently dragged
round.

(3) Mallet's second theory, which attributes the twisting to a
second shock, affecting the object twisted while it is
tilted on one edge or corner by the first. The second
shock being oblique to the first,causes the object to
rotate on the edge on which it is tilted.

(4) Gray's theory, which attributes the rotation to the centre

( 217 )

E PRA TAN tv

218 OLDHAM: GREAT EARTHQUAKE OF 1897.

of gravity, and the corner on which the object is tilted
not being in the same line with the direction of the shock
and to the object, being consequently swung round on
that angle,
In discussing these theories will it be convenient to take them in
the reverse order to that in which they are stated above,
Commencing with Gray’s, it will be noticed that, whereas this
demands that the rotation should be round one of the corners of the
object rotated, a study of the plans on Plates XXIV-XXVII will show
that such was not the case in a single instance. Besides this it fails
to account for one of a number of monuments all similarly oriented
being rotated positively while all the rest were turned negatively.!
Mallet’s second theory could be made to account for these cases,
but it demands that the rotation should be
| round one of the edges of the surface at
iC which it took place. Now every case of
rotation may be explained by a single
revolution round a fixed centre, and the
radius of the circle in which the centre of
gravity may be supposed to have revolved,
in order to bring it from its old to its new
position, is easily calculated.
To determine this let A and B, fig. 21,
be the old and new positions of the centre

Fig. 21.

of gravity. Then the required centre must be somewhere on the
line which bisects A B at right angles, for it must be equidistant
from A and B. But in moving the centre of gravity from A to B,
the rotated object has also turned on its axis through an angle which
we will call. If then two lines are drawn from a point C so that

1 The plans on Plate XXXVIII appear at first sight to favour this hypothesis. The monu-
ments, however, are so low in proportíon to their base that it is very doubtful whether any tilting
took place, and the displacement was probably due to the vertical components of the shock,
and was ofthe nature of a projection. It is possible, however, that these cases may be due
to a single oscillation of greater power and extent than tbe rest, for the possibility of the
action involved in Gray’s explanation cannot be doubted. Though it cannot be accepted as
a complete explanation,

( 218 )

ROTATION OF PILLARS. 219

each may make an angle of 4 with the central line, and cut the
points A and B, respectively, we have an isosceles triangle, whose
apical angle is, and whose two equal sides are radii of the Circle in
which the centre of gravity may be assumed to have moved, to
enable it to pass from its old to its new position, while turning
through the angle AC B=09. If this radius be represented by 7, and
the distance A B by d, then

deve 8
7—-—--smn ~=
2 2

According to Mallet's theory, » should equal the distance of the
centre of gravity from the point or edge on which the object rotated,
that is to say it should be equal to, or lie somewhere between, the
semidiameter and semidiagonal, and should not materially differ from
them in excess or defect.

In applying this test it will be necessary to reject those cases where
the apparent shifting of the centre of gravity and the angle of rotation
are both small. Neither can be measured with very great accuracy,
and when either is small a very slight absolute error would bea very
large one proportionately, and lead to great errors in the result.
Applying the test to those cases where it may be expected to give a
reasonably accurate result, we have, in the case of Inglis’ monument
at Chhatak, y=27°8 in., while the semidiameter is 52 in. In the case
of the monument No, 2 on Plate XXI, fig. t, 7-3 in., while the semi-
diameter is gin. In the case of the W pillar E gate, and the E pillar
W gate of the telegraph signallers’ quarters at Gauhati » =5 in.
and 24 in. respectively, while the semidiameter is 124 in. in both
cases.

From this it will be seen that there is no correspondence between
the calculated radius of revolution and that required by the theory.

Mallet's first theory, though undoubtedly an explanation, is insuffi-
cient as it does not account for the fact that a number of similar objects
are similarly rotated in the same neighbourhood; nor does it seem
probable that there would be so greata divergence between the
positions of the centre of gravity and the centre of resistance, and what

( 219)

80 OLDHAM: GREAT EARTHQUAKE OF 1807.

is more important, so complete a check to lateral movement at the
latter, as would account for the observed rotations in many cáses.

There remains but the first mentioned explanation, that of
vorticose motion, and with regard to this it seems that the objections
raised are based on an unnecessarily restricted interpretation of the
term, It has been supposed necessarily to imply a rotation round an
axis, by which a similar rotary or top like movement is given to the
objects rotated.

This is evident in Darwin's objection to the theory,and is even
more forcibly put by Mallet, who wrote ! :—

** The sagacity of Darwin showed him that the vorticose hypothesis was im-
probable, and that in order to its being at all tenable, a separate vertex must be
admitted for every separate stone found twisted, the axis of rotation of the vortex
having been coincident with that of the stone. Besides this paramount improba-
bility, therefore, a little further reflection would have led either Lyell or Darwin to
estimate the inconceivable angular velocity of motion at the extremity of the
radius of one of these vortices, even if assumed at no more than a few hundred

feet, necessary in order that its velocity within à few inches of the centre should
beso great as to wrench DUE of its mortared bed, and twist a block of masonry

by merely its own inertia."

Here we have the assumption that what is called vorticose
motion must necessari be a twisting round an axis, and the
absurdity to which it leads, both clearly expressed.

But though, interpreted in this way, the term vorticose leads to an
untenable hypothesis it must not be forgotten that, while popular
interpretations of natural phenomena are frequently incorrect, they
almost invariably represent a real substratum of fact. That the |
vorticose should have been for generations regarded asa distinct
type of earthquake movement points to there being a real difference
between it and the types which are classified as ‘horizontal,’ *undu-
lating’ or * palpitating.” Abandoning the restricted interpretation, it
is not difficult to see that it might well be applied to movement of a
totally different character, that is toa movement of the waves

1 Jour. Geol, Soc. Dublin, 111, 138 (1845).
( 220 )

ROTATION OF PILLARS. 221

particle in a more or less circular course. Here there would be no
angular rotation, but over a larger or smaller area the whole ground
would be caused to move in small circles, and a pillar or similar object
fractured from its base, would be caused to tilt and roll round on its
edge and so rotate with regard to its original position.

It does not seem, however, that the term ‘ vorticose’ is necessa-
rily restricted to so simple a form of movement as this, for 1
find in Mr, Mallet's account of the Neapolitan earthquake a state-
ment that the Syndic of Padula, was of opinion that the shock was
* vorticose ' or at least in various directions transverse to the main
one and so close together in time that it was impossible to regard the
earthquake as other than a prolonged and irregular succession of
oscillations, lasting several seconds.! From this it is evident that the
term vorticose is applied to shocks in which no definite direction of
movement can be noticed, but where the apparent direction seems
to vary or revolve. This interpretation finds its explanation in
the diagrams of the seismographs which have been invented and
taken into use in Japan.

Fig. 22. Diagram of vorticose motion.
To take the simplest case first, suppose the movement of the
wave particle to be circular, neglecting for the moment any ` move:

1 Great Neapolitan Earthquake, 1857, I, p. 368.
| ( 221 )

222 OLDHAM : GREAT EARTHQUAKE OF 1897.

ment there may be in a vertical direction. In fig. 22 let the outer
circle represent the ground plan of a cylindrical column, and the
inner circle the path of the wave particle, that is to say, the track of
a point on the ground, all other points in the neighbourhood describ-
ing similar circles. At N the movement of the ground is eastwards,
causing the column to tilt upon its western edge; when the base
motion has progressed so that the particle at N has reached E, its
motion is southwards and the column will be tilted on to its northern
edge. But it had been tilted up on its western edge, consequently in

being thrown on toits northern it will be revolved slightly in a
negative direction. So when the point N has reached S, the move-

ment will be west and the column tilted on to its eastern edge and
again revolved slightly in a negative direction, and so on, For
simplicity of treatment an abrupt change of motion from east to south
and from south to west has been supposed, but the effect of a
gradual change would be the same, a continual tendency to alter the
point on which the column was tilted and to cause it to rotate in the
opposite direction to the course of the wave particle.

It is also to be noticed that the centre of the column will i
scribe a spiral line round its original position, as each change of tilt
will carry it a little further out, but its final position will be closer
to its original one than would be the case, had the same amount
of twisting taken place by asingle tilt on one edge, and a si-
multaneous sidelong displacement. In other words the virtual
radius of revolution would be less than the semidiameter ;
being thus in better agreement with facts than either the third or
fourth hypothesis.

There is another type of earthquake motion, more common than
that required by the preceding explanation, which may equally well
give rise to rotation, This form may be diagrammatically represented
in fig. 23, supposing the movement to start from 1, then in the
path from 1 to 2 the column will be left behind and caused to
tilt towards 1, while the change of direction between 1 and 2 may

(.222 )

ROTATION OF PILLARS. 223

give it a slight positive rotation. As soon, however, as the direction

Fig. 23. Diagram of'rotation by vorticose motion.

of the movement changes and the particle begins to return from 2 to
3, a much more powerful tendency to rotate the column ina negative
direction round 1 will be set up, and the column at the same time
tilted over on its edge towards 2, The same thing will happen when
the particle returns from 3 towards 4 and so on, each successive tilt
being accompanied by a further twist in a negative direction. As
before, the distance of the centre of the column from its original

position will be less than it had there been merely a single tilt and
twist.!

Of these two explanations the latter seems more in conformity
with the facts observed than the former. The first would require
that the crushing of the edges should be tolerably uniform all round,

1 This chapter was written and set up in type before 1 was aware of a paper by Mr. C.
Davison on the theory of vorticose earthquake shocks (Geol. Mag., dec 2, IX, 257 — 266, 1882),
which would otherwise have beef referred to in the text. He explains the vorticose shock
much in the same way as that adopted in the text, i.e. as a movement to and fro in a gradually
changing direction, and points out how such must necessarily result with a focus of any size, in
the neighbourhood of the epicentre, from the fact that the impulses reaching a spot at succes-
sive intervals of time come from different parts of the focus and hence from different directions.
'The manner in which such a form of movement leads to the rotation of solid objects is not
discussed, but it is held (p. 264) that the objection to the vorticose theory urged Mr. Darwin (in
1839) that under every object so turned there must have been a separate vorticose movement,
loses its force on the view there presented. |t is evident, however; from the extracts quoted
in this chapter that the vorticose motion, as understood by Darwin and Mallet, was something
very different to that whose possibility has been demonstrated by Mr, Davison, whose existence
has been proved vy the researches of Prof. Milne and others in Japan, and whose actuality is

adopted in this memoir as a satisfactory means of explaining the rotation of objects by an
earthquake,

( 223 3

224 OLDHAM: GREAT EARTHQUAKE OF 1897.

while it was noticeable that this was not the case. As a rule, only
certain parts showed signs of crushing, while others were left sharp-
edged and intact. This might, of course, be due to unequal powers
of resistance of different parts of the borders of the plane of fracture
and twisting, but is more probably due to differences in the pressure
brought to bear on them.

In the foregoing passages I have, for simplicity's sake, assumed
a circular section of the column twisted, but the explanation is
equally applicable to objects of square or angular section, In this
case the first tilt would necessarily be on one side or one angle,
and Gray’s principle would come into play, but all succeeding tilts
would take place along one of the sides of the lower part of the
column, or at one of the points of intersection of the sides of the
lower and upper portion, after the former had been twisted from its
original position. In this way may be explained the fact that
some of the corners show quite clear, cut and uncrushed fractures,
while on the sides close by there may be very palpable signs
of the effect of great pressure. |

It may be noticed that the explanation which has been proposed
is but an amplification of Mallet's second explanation. Instead of
only two shocks it postulates a series of successive shocks in suc-
cessively varying directions, and by regarding the rotation as due,
not to a single twist round a single centre, but to the accumulated
effect of a succession of small twists round a succession of different
centres, it allows a considerable rotation to be accompanied by a
much smaller shifting of the centre of gravity than could be the case
if the same amount of rotation was due to a single effort of twisting.

If we modify Gray's explanation in so far as to regard the rota-
tion as due not to a single tilt, but to a series of successive tilts on
opposite sides of the base, we might get a large rotation accompanied
by a small shifting of the centre of gravity, but it fails, even with this
modification to explain how two similarly oriented objects close
together may be rotated in opposite directions, We heve two cases

( 224 )

ROTATION OF PILLARS, 225

of this: one of the two pillars at Inglis bungalow, Cherrapunji, anda
more striking one in the old cemetery at Gauhati. In both these
cases it is to be noticed that the height of the centre of gravity of
the displaced portion above the plane of fracture is smaller in the -
case of one direction of rotation than of the other.

Now, in the explanation it was assumed that the column once
tilted on to 1 would remain so till the motion of the wave particle
ceased at 2, and the return movement from 2 to 3 set in. This
would generally be the case, but the smaller the ¡height of the
displaced portion, the more quickly will it recover its position
of stability, and the less easily will it be displaced, for the same
section of base. As a consequence there comes a stage at
which, as the height of the displaced part diminishes, it will only
be tilted as the wave particle attains its maximum acceleration, which
may be put at about one-quarter of its path from 1 to 2. Before
the wave particle has reached 2, the displaced part will have fallen
down into its base again, retaining what positive rotation it may have
acquired owing to the gradual change of direction of the path of the
wave particle, or to the direction of this with regard to the diagonals
of the rotated portion and by a repetition of this process acquire a
gradually increasing angular displacement in a positive direction. :

But there would almost certainly be more than this. Instead of
merely falling into its base the displaced part would. probably be
tilted up, by its momentum, in the direction of 2, or of somewhat
towards 5 from 2, before the wave particle reached 2. It would then
be subjected to a displacement due to the movement from 2 to 3,
that is to say, it would be rotated in a positive direction, or the
opposite to that which would be impressed on it if it had been tilted
towards 1 while being acted on by the movement from 2 to 3.

From this we see that the direction in which the displaced part
will be rotated depends not only on the relation between the width
of, and the height of its centre of gravity from, its base, but also
on its absolute dimensions. In other words, it depends on the quick-
ness of recovery of the displaced portion when tilted. It has not

Q ( 225 )

226 OLDHAM : GREAT EARTHQUAKE OF 1897.

been possible, for want of a sufficient number of suitable objects, to
fully test this hypothesis, but the cases of the Gauhati old cemetery
and of the two pillars at Cherrapunji are explicable by this, and
by none of the other hypotheses.

We have consequently an hypothesis which explains—

(1) Why a number of similarly oriented neighbouring objects are
generally twisted in the same direction.

(2) Why in certain cases we may have an object twisted in one
direction, while others similarly oriented but of different proportions
are twisted in the opposite direction. |

(3) Why the displacement of the centre of the displaced portion
from its original position indicates a centre of revolution which does
not coincide with the outer edge of the surface of displacement.

It explains, in short, all the observed facts, while it involves nothing
which is inconsistent with them. We may take it, then, that the rotation
of objects is the consequent of a rotary movement of the ground, which
may take the form of either a more or less circular path of the wave
particle, or a succession of backward and forward movements, each of
which takes place in a fresh direction different to the previous one.

Other explanations are not, however, excluded by this, Gray’s is
indubitably a possible cause of twisting, while the statement of Michele
Sarconi, that of the two similar pillars at either end of the facade
of the Certosa di S. Bruno one was turned from right to left and the
other from left to right", points to the conclusion that Mallet’s first the
explanation is good in some cases. All that can be claimed is that
the existence of a movement of the ground which may be called
verticose does exist, and that it is a cause of the rotation of objects
by earthquakes; probably it is usually the correct explanation,
though in individual cases the twisting may be due to other causes

and not inconsistent with a rectilinear movement of the wave
particle.

: [storia de Fenomeni del Tremoto avvenuto nelle Calabria, &c., Naples 1784, page 62.

( 226 )

THE UNFELT EARTHQUAKE, 227

CHAPTER XV.—THE UNFELT EARTHQUAKE.

Beyond the limits of the area over which the earthquake was re-
cognised without instrumental aid, its effects were impressed on the
records not only of instruments which have beer set up for the special
purpose of detecting the imperceptible tremors and undulations which
traverse more orless the whole of the earth, after great earthquakes,
but also on those of self-recording magnetic and electric instruments.
After the earthquake every endeavour was made to obtain as
complete a set of these records as possible, and I have to acknowledge
the readiness with which information was rendered, and even
photographs and laborious tracings of the records, made and sent.!

The unfelt disturbances due to distant earthquakes are commonly
referred to as microseismic, but the term is hardly applicable to
undulations which, in the case of this earthquake, had a length
of 34 miles and caused the ground to rise and fall 20 inches? Instead of
microseismic I would introduce the term cryptoseismic as more
suitable; it implies nothing except that the movement is not obvious,
and is useful to distinguish between the unfelt'movements due to a
distant earthquake, and the movements due to an earthquake of so
small an order of magnitude that it never becomes sensible, except
to very delicate instruments ; to these the term microseismic might
well be restricted. |

The records vary much in value according to the nature of the
instruments employed. In many cases these were in reality too deli-

1 A very detailed account of the records of this earthquake, obtained in Europe, has been
compiled by Dr. G. Agamennone and published in the Boll. Soc. Sismol, Ital., 111, pt.ii, pp. 249
—293. In this details are given of some records which I was unable to obtain. The account
of those referred to is also more detailed and lengthy than that printed in this chapter.
In some instances the times differ slightly but no alteration has been made in those printed,
which are the times as originally communicated to me, as the changes were too slight to justify
the labour of recalculation. For the same reason the details regarding records at stations
from which 1 had not previously obtained information have not been incorporated in the
chapter, with a few exceptions specifically referred to,

? These figures are taken from Prof. Agamennone, Boll. Soc. Sismol. Ital., IV, pt. i, 66 (1893).
They are intended to apply specially to Italy but are equally truelfor the whole of Europe.

Q 2 ( 227 )

28 OLDHAM: GREAT EARTHQUAKE OF 1897,

cate, and, under the influence of so exceptionally great a disturbance
as was set up by this earthquake, the record of all but the beginning
and end was lost. In other cases the rate of travel of the recording
surface was so slow that the time cannot be accurately determined,
From the magnetic observatories the replies were mostly negative,
and in other cases the disturbance was of sucha nature as to preclude
the possibility of exact time determinations. In the following pages
it has not been considered necessary to refer to these cases, and only
those records which have given tolerably trustworthy determinations
of time are referred to, The times as given in the original record
have in every case been converted into Greenwich mean time,

The records of the cryptoseismic disturbance fall naturally into two
classes,—those of instruments specially and solely intended to record
disturbances of the ground, and those of magnetic or electric instru-
ments, not intended to record movements of the ground on which
they are set up, though giving evidence of earth movements by an
irregularity or break in their record. Foremost among those of the
first class I must place the valuable series of records for which I am
indebted to Prof. Grablowitz, and not only for the records but
also for a discussion and comparison of them, from which the nature
of the series of disturbances is clearly deduced.

Proceeding in the first place to the description of the facts re-
corded, I shall afterwards discuss their meaning and interpretation.

Casamicciola (Ischia).—I am indebted to Prof. G. Grablowitz,
Director of the well equipped geodynamic observatory at Casa-
micciola, for a most interesting account of the effects of the earth-

quake on his instruments. These are—

(1) A pair of horizontal pendula, arranged so as to complete an electric circuit,
on the smallest disturbance, by which an alarm is sounded, a time-signal impressed
on the recerd of No. (3), and No. (4) set in motion. It acted at 11h. 19m. 3s.
G SLT.

(2) The “vasca sismica; a circular cistern 1°56 metres diameter and 1:10 m. deep
excavated in the basement of the observatory and filled with water to the depth
of ım. On this there is a float tem, diameter, the movements of which are regis-
tered, with an amplification of 92 times, on a recording surface moving at the rate
of about 33 cm. per hour.

( 228 )

THE UNFELT EARTHQUAKE, 229

(3) Consists of a pendulum of 1m. length and a weight of ro kgm. It is in-
tended to register sensible earthquake, and the record, which is started by (1), only
lasts for one minute.

(4) A simiar instrument, but with a weight of 20 kgm. and a continuous record.
The recording surface moves at the rate of 10 cm. per hour, and on it, besides the,
record of this instrument, the time at which (1) closes circuit is also marked.

(5) A pair of delicate spirit levels, set N-S and E-W, which are directly
observed by eye.

(6) A spiral spring of 7m. length supporting a weight of 150 grammes,

(7) Self-registering geodynamic levels. These consist each of a pair of cylin-
drical receptacles 30 cm. diameter and 25 cm. high, connected by a tube some 2 m,
long and 15cm. diameter. There are two, set N-S and E-W, and the movement
of the water surface is recorded, with an amplification of 50 times, on a surface
moving 11 mm. per hour.

(8) A pair of horizontal pendula, with weight of 12 kg. recording their moves
ments with an amplification of 1 to 8 on a smoked paper moving at the rate of
30 cm. per hour. The period of vibration ofthe N-S pendulum was 13'5s., and of
the E-W one 13’0s., on the 12th June.

(9) A similar but improved instrument with three pendula of a weight of 3kg*
oriented at 60^ to each other.

Of these instruments Nos. 1 to 6 are set up at the Grade Sentinella, Casamic-
cola, and 7 to 9 at the Port of Ischia.

All these instruments acted on the 12th June. No. (1) gave the
alarm at 11h, 19m., 3s. Greenwich mean time.

No. (2) shows a sudden commencement of rapid oscillations at 11h.
17m. 135., which give place to slow oscillations of a complete period
of about 20s. at 11h. 42m. 33s. These attain their maximum at
about 11h. 47m., whilst for a couple of hours the diagram shows a
much smaller but continuous movement.

It may be noticed that both the first rapid oscillations and the
subsequent slower ones show a considerably greater amplitude in
the E-W component than in the N-S ; specially so in the case of the
former. On the copy of the diagram of this instrument, for which I
am indebted to Prof, Grablowitz, I find that the E-W component
is three times the N-S in the case of the first, and one and a half
times in the case of the second.

The record of (3) is of little interest as it is limited to a period of
one minute only. No. (4) shows a slight thickening of the line of the
E-W component for a few minutes, but there are records of repeated
clesures of the circuit by (1) for nearly an hour.

( 229 )

230 OLDHAM: GREAT EARTHQUAKE OF 1897.

The instrument (5) was observed immediately after the sounding
of the alarm. The E-W level was specially disturbed and was making
oscillations of not less than 1” of arc during the first phase.
The movement had almost ceased when, at 11h. 42m. the oscillations -
became considerable, reaching, at 11h. 47m., an excursion of 4'5” of
arc from E to W, and 3'o" from N to S, with a complete period of
15s. to 20s. After 12h. the oscillations diminished and at 12 h. 20 m.
there was only a movement of about 0'3" on each level.

The spring (6) showed, after 11%h., vertical oscillations of not
more than 2 mm.

Fairly satisfactory diagrams were obtained from (7) and (9),
considering the slowness of travel of the record,

No. (8) gave by far the clearest and most prolonged diagram, for
as late as 22h. there were still oscillations, showing that the earth only
came to rest after eleven hours ; it is probable that these were the
effect of later shocks, registered by this, but not by the other
instruments.

From the combination of these records, and especially those of
(2) and (8) we may obtain with certainty the following results :—

H M. S.
Commencement of the rapid vibrations s qu 17 13
Maximum of ditto antem 19 3
Commencement of the slower movement ! Tu 26 o
Commencement of large undulations . 12 42 33
Maximum of the same > ^ 5 SET 47 O
Decrease of the same . ` . 5 e IE 57 o
End of phase of sensible movement : 3و‎ o o

With reference to the above results Prof. Grablowitz points
out that there are two principal phases deserving special examina-
tion, the maximumof rapid motion, at 11h. 19m., and the maximum of
slow movement, of a complete period of about 20seconds, at 11h. 47m.

The amplitude of movement in the two phases varies much in the
different instruments, but if their different conditions are considered,

' Maximum of this phase 11h. 27m. 40s. ; Boll. Soc. Sismol. Ital, HI, pt. ii, p. 26.
The time of commencement is given as 11h, 25m. 48s.

( 230 )

THE UNFELT EARTHQUAKE, 231

as well as the various instrumental amplifications, the results are
such as to lead to the conclusion that the problem of the seismic
motion requires a different solution in each case.

In the first phase the measured amplitude, after allowing for the
respective instrumental amplifications, is as follows :—

RAPID MOVEMENT SLOW MOVEMENT
OF IIH. 19M. OF IIH. 47M.

سس سحح و | ms‏ ——— ہس | AS‏ ————

mm mm mm mm mm mm
(2) Vasca sismica . > ə el 0°20 0°05 0'21 0'038 | 0'027 | 0'047
(3) Seismometograph . . el o'1o 0°00 0' Ce) 0'040 | 0'000 | o'04o
(5) Spirit levels e . ` sl. 200 ? 2'0 g'o Go 11*0
(7) Water levels . . . «| Go | 0'014 | 0042 | o'olo | o'oo0 | o'oro

(8) Horizontal pendula . . «| o'r9 0'12 0'23 5'50 285 6'20

—— M—ÓM—————————M———————————Á

These linear amplitudes may be converted into angular values, |
on the basis of the equivalent lengths of a single pendulum to which
the instruments correspond, which is given in the first column of the
table printed below.

MA. Pe

Equivalent ANGULAR VALUES,
length of

رہہ ae‏ یں او ری ہو رھش pendulum.‏

11h. 19m. 11h. 47m.
‘Metres. a s
(2) 4 ۰ è © o a e 0°75 57'8 13'0
(3) . . . . ۰ H ۰ 100 20'8 82
ہے کر وک ںےہ رد اتہب‎ 412'00 vo 55
A را‎ s ` e : 1'10 79 |»
E. W. € B L] LJ D 41'7 ; à Lo 272
(8) MiS Ta A * e à 452 o'6 13'0
Resultant . . . e eene ; I'I 30'1

232 OLDHAM: GREAT EARTHQUAKE OF 1897.

Of the two modes of interpreting the records of the first phase,
that of regarding them as representing angular movements presents -
enormous differences between the results obtained from the different
instruments, while, if they are regarded as representing linear dis-
placements, they merely exhibit differences explicable by the varying
natural and mechanical sensitiveness of the instruments, On the
other hand, the measures of the second phase present very great
differences of linear, but not so great of the angular values, in
which they follow an evident law, that the values are higher as the
instrument from which they are derived is mechanically freer from
friction or resistance.

The interpretation of this may be that the earlier disturbances
are, at any rate largely, the effect of inertia, due to horizontal move-
ment (condensational waves), while, those at the later period were
due principally to an actual tilting of the surface, consequent on
movement of the nature of an undulation (distortional waves).

As a consequence, Prof. Grablowitz considers that the maximum
inclination of the surface at Ischia was 2”, that registered by the water
level, the only instrument in which adhesion to the sides could com-
pletely, or nearly so, extinguish the disturbance of the horizontality
of the surface of the liquid by waves of such period, and the effects
of inertia felt by other instruments; while the excess observed on
the other instruments, even in the spirit levels, is probably to be
ascribed to inertia consequent on horizontal displacement. The
oscillation of the spiral spring as observed seems to be sufficiently
confirmatory of this hypothesis, as it agrees in time with the arrival
of the undulatory waves, which were accompanied by a considerable
amount of vertical movement.

I have reproduced the substance of Prof. Grablowitz’s account of
his instruments and the conclusions he drew from their records, not
merely on account of their intrinsic interest, but also on account of
the assistance they gave in the interpretation of the records from

1 These are different to the distortional Plane waves transmitted ¢hrough a solid.

( 232 )

THE UNFELT EARTHQUAKE. 233

other observations. The two important points are (1) the sudden
commencement of the nearly horizontal movements of short period,
and their subsequent gradual dying out, and (2) the gradual coming
in of the slower movements accompanied by an actual tilting of the
surface, in which the horizontal was combined with vertical motion.
These increase to a maximum and then diminish again, but continue
for about an hour and a half. As will be seen when the records as a
whole come to be discussed, Prof, Grablowitz's second is really a
third phase, as there is another recognisable, intermediate in time,
between the two discussed by him.

Rome.—R. Osservatorio Geodinamico di Rocca di Papa.—
I am indebted to Dr. Cancani for photographs of the records
obtained in this observatory from (1) a horizontal pendulum instru-
ment of which the N-S pendulum has a period of 14s. and the E-W
of 12s.; and (2) of a simple pendulum of 15 metres length anda
bob weighing 250 kilogrammes.

The photograph of the trace of the vertical pendulum does not
contain the commencement or the end of the disturbance. Slower
undulations with an increased amplitude, marking the advent of the
second phase, commence about 11h, 25m. 58s. and the long undula-
tions at 11h. 41m. 50s., attaining their maximum at 11h. 47 m. 8s.
As at Ischia, the largest movement is in the E-W direction.

A copy of the diagram of the horizontal pendula is reproduced on
Plate XLI. From this it will be seen that the commencement of the
disturbance is very distinct on the E-W component at 11h, 17m. 40s.
The pen of the N-S component was not writing properly and its
trace is very faint till 11h. 27m. 51s,, when the line becomes distinct
once more; in spite of the faintness of the trace the commencement
of the disturbance can be recognised. The second phase is not
well marked by this instrument, but its commencement can be recog-
nised by anincrease of amplitude and more open character of the
waves at 1th. 25m. 54s. The great undulations are very conspicuous
' on this diagram, commencing at 11h, 41m. 40s. and attaining their

( 233 )

234 OLDHAM: GREAT EARTHQUAKE OF 1897.

maximum on the N-S pendulum at 11h, 47m 8s. The maximum on the
E-W component is somewhat later, but the maximum of the resultant
of the two coincides very closely with that on the N-S component.
'The trace continues disturbed till 14h.

Rome.— Ufficio centrale di Meteorologia e di Geodinamica,—
A very detailed account of the records obtained at this observatory
is published in the Boll. Soc. Sismol. Ital, II, pt. ii, pp. 253-267,
from which the following has been abstracted :— |

Sismometografo medio (8m., 100 kg., amplification 10 times) سس‎
Commencement abrupt at 11h. 17m. 5s.+5s., maximum displacement
2:5 mm, NW-SE, 3'0 mm, NE-SW; resultant nearly E-W, After
coming to rest the disturbance due to the second phase commenced
between 1th. 24m. and t1h. 24m. 205., the maximum displacement
being 3 mm. shortly after the latter time. The great undulations
began at 11h. 43m. 30s. and reached their maximum of romm. at 11h,
47m. 30s. The disturbance ended at 12h. 5m.

Sismometografo grande (16m., 200kg , amplification 12 times).—
Commencement at 11h. 17m, 20. + 5s., maximum of first phase 5mm.
at 11h. 20m. 20s., decreasing to almost extinction, The second
phase commenced at 11b. 25m, 20s. reaching a maximum at 12h.
28m. 15s. on the NE-SW component, and at 11h. 27m. 30s. on
the NW-SE. The great undulations commenced at 11h, 43m. 8s.
and reached their maximum at 11h, 46m. 58s. or shortly after.

Padua. —In the University of Padua two microseismographs have -
been set up by Prof, Vicentini; one a simple pendulum of 1'5 metres
length with a bob weighing 100 kgm. ; the other a simple pendulum
of 10 metres length and 400 kgm, weight." A reproduction and dis-
cussion of the diagram obtained from the first of these is given in a
paper by Dr. M, Baratta.? As at Ischia, the first movements were
of a rapid to and fro nature ; they commence somewhat abruptly at

1 For a detailed description and illustrations of these instruments, see a paper by
Dr. Giulio Pacher, Atti del R. Instituto Veneto de scienze lettere ed arte, 7th series,

VIII, 1896-97, and also in Boll, Soc. Sismol, Ital., III, pt. i, 65-131 (1197).
2 Boll. Soc. Geog. ltal., 1897, fasc, viii.

( 234 )

THE UNFELT EARTHQUAKE. 235

11h. 16m. 47s. G. M. T., and have a maximum between 11h, 22m, and
11h. 24'5m. The second phase commences at 11h. 26°75 m. and
reaches a maximum at 11h. 28:25 m. Dr, Baratta points out that
after 11h. 35m. there is a marked change in the character of the
curve, and instead of the oscillations of the pendulum being made
evenly on either side of a central line, they are made on either side
of a sinuous line such as would be due to a slow tilting of the whole
instrument backwards and forwards. The well marked slow undu-
lations set in at about 11h. 42m. 30s. and attain their maximum
at 11h. 47m. 10s. The disturbance continued till 13h. 30m, G. M. T.

The long pendulum of 10 metres gave a diagram showing the
commencement of the disturbance at 11h. 17m. and its cessation
at about 15h. G. M. T. The greater duration of the disturbance
on this instrument is due to its greater sensitiveness.

The horizontal displacement of the pendulum bob during the first
phase amounts to as much as ‘5 mm., and between 11h, 27 and 11h, 33m.
the maximum displacement attained is ‘47mm., or nearly as much,

The displacement at 11h. 47m., if wholely due to tilting of the
instrument, shows a tilt of 22” of arc from the vertical.

Catania.—I have received from Prof. A. Ricco, Director of the
Royal Observatory, three copies of the records of his two seismo-
graphs ; one of these only gives a record of one minute duration and
the diagram is consequently not of great interest. The other has a
continuous record; it is,like most of the Italian instruments, a
simple pendulum, of 25°30 metres length and 200 kg. weight, whose
movement is magnified 12'5 times and recorded on a paper moving
at the rate of 0'6 metres per hour." The components are registered in
NE-SW and NW-SE directions.

The disturbance of the trace, which is reproduced on Plate XLI,
commences at 11h. 17m. 23s. the total displacement of the weight
being '3 mm. from the vertical. The commencement of the second

1 A description of this instrument, written by Prof, Ricco, will be found inthe Atti dell.
Ac. Gioenia de Sci. Nat. in Catani, 4th series, X,

(235 )

p

26 OLDHAM: GREAT EARTHQUAKE OF 1897.

phase is well marked at 11h, 25m. 44s. on the NE-SW component,
and at 11h.25m. 55s. on the NW-SE component, the character
of the trace becoming more open, and the travel of the pendulum
bob rises to rt mm. the character of the curve indicating oscil-
lations of the normal period of the pendulum, 5*2s., superimposed
on undulations of longer period. These correspond to the violent
oscillations of the Padua pendulum between 11h. 26m, and 11h. ۰
Converted into angular measurement, the excursion of the pendulum
corresponds to a tilt of 10” of arc from the vertical. The larger
undulations set in about 11h, 45m.; their maximum is at 11h. 48m,,
but as the components are NE-SW and NW-SE instead of N.S and
E-W, the diagram cannot be readily compared with the others, and
it is possible that the apparently late beginning of these is due
to the E-W component not showing up as well as it does in the
other diagrams. The disturbance lasted till about 14 h.

Siena, —I am indebted to Prof. Silvio Lussana, of the University
of Siena, for a photograph of the diagram made by his instrument.
This closely resembles on a small scale that of Prof, Vicentini's
instrument at Padua. Disturbance commenced at 11h, 17m. G. M. T,
rapidly attained a maximum at 11h. 19$ m. and then died down.
The second phase is marked by an increased displacement from
26m. to 28m., with a maximum at 27m., but instead of being
nearly equal to the original displacement, it does not amount to
more than one-sixth of it. The trace after this shows a number of
minute undulations, and at 46m. they increase in size, attaining
their maximum at 48 m., and then diminishing once more. I have no
information as to the total duration of the disturbance,

Pavia.—From Dr. Emilio Oddone, of the Royal Meteorological
and Geodynamical Observatory at Pavia, I have received a tracing
of the record made by his microseismograph, a simple pendulum
of 4'5m., length, and gokg weight; the movement of the bob
is magnified ten times, and recorded on a paper moving about 10 cm.

per hour.

( 236 )

THE UNFELT EARTHQUAKE. 237

The disturbance commenced at rih. 18m. 17s. G. M. T., and
at 45m. 43s. there was again an increase in the travel of the
recorder, reaching its maximum at about 11h, 48m,, the distur-
bance of the trace ceasing at 12h. gm. The second phase appears
to be marked by a large increase in the travel of the recorder of the
E.-W, component at 11h. 27m., but the trace is not open enough to
allow this to be decided with certainty.

On this trace it must be noticed that the N-S component is
about double the E-W in the first phase. There is a maximum at
about 11h. 20m. and another large one at 11h. 23m. The other
increase in the travel ofthe pointer after 45m. 43s. evidently corre-
sponds to the large undulations of other records. At Pavia the period
of these undulations was about 30s. ; in the first phase the oscilla-
tions were very rapid.

Grenoble.—A microseismograph here recorded the earthquake.
Beyond the fact that the disturbance commenced at 11h. 28m. 26s.,
Paris time, or 11h. 19m. 5s. G. M. T.!, I have no information as to
either the nature of the instrument or of the remainder of the record.

Strassburg.—Dr. Gerland informs me that his instrument, con-
sisting of three horizontal pendula set at angles of 120° with each
other, was violently affected by the earthquake. The disturbance
began at 11h. 18m. 32s. G. M. T. without any tremors, with sudden
and very great oscillations, of 14 cm. amplitude; these suddenly
ceased at 12h. 45m., but were succeeded by a very long series of
diminishing tremor-like oscillations, dying out at 14h. 30m. G. M. T.

Potsdam.—4A horizontal pendulum, with a boom 20cm. long and
a weight of 10 grammes, supported by a quartz fibre rising at an
angle of 45?, was set in motion at 12h. 11m, Potsdam time, which is

1 This time was communicated to me by Dr. Früh of Zurich. In Symons’ Monthly
Magazine, XXXII, 92 (1897) the time is given, in a note by Prof. J. Milne, F.R.S., as
11h. 28m. os. According to Dr. Früh's communication the time is Paris mean time and requires
a correction of 9 m. 21s. This also brings the time more into accordance with that recorded
elsewhere. The time is givenas 12h, 19m 5s. T. E. C., equivalent to 11h. 19m. 5s. G. M. T. by
Dr. Agamennone, Boll. Soc. Sismol. Ital, UN, pt. ii, 292.

(237)

28 OLDHAM: GREAT EARTHQUAKE OF 1897.

52m. 15s. in advance of Greenwich. The G, M. T. of commencement
is, therefore, 11h. 18m. 45s. The times can be relied on to half
a minute, At oh, 37m. G. M. T. the horizontal pendulum was
thrown out of gear and the record ceased.

Newport, I. W.—At Shide, near Newport, in the Isle of Wight,
Prof. J. Milne, F.R.S., has established a seismograph of his own
design. The record is photographic, and unfortunately a change of
paper was being made when the disturbance commenced ; the record
of the commencement of this earthquake was consequently lost,
but the end of the disturbance was determined as 14h. 32m. 51s.
The largest angular tilt recorded was 5" of arc, but it may have
been greater in the part of the record which was lost. A horizontal
pendulum with a boom 4 ft. long gave the period of thelarge oscil-
lation as 15s. i

Edinburgh.—At the Royal Observatory, Edinburgh, is a bifilar
pendulum, an instrument consisting. of a mirror hung from two
threads of unequal length whose upper ends are fastened nearly
vertically above each other. As long as the two points of suspension
and the two points of attachment remain in the same vertical
plane, the mirror remains at rest, but as soon as it is tilted in the
slightest degree in any direction, except that of this plane, the mirror
swings round till the four points once more lie in the same vertical
plane, The mirror is immersed in petroleum to prevent its swinging
too freely, and the tilting to which the instrument is subjected is
registered by means of a ray of light reflected off the mirror on to a
piece of sensitive paper. |

. A description of the effect of the earthquake on this instrument
has been published by Mr. T. Heath!, along with a reproduction of the
photographic record. ‘The first indication of disturbance is at
11h. 18m,, when very slight preliminary tremors are noticable:
at 11h. 28m. more violent oscillations set in and lasted till 12h.
33m. After this, there is a slight but distinct disturbance at

1 Proc, Roy. Soc., Edin,, 1897, pP. 481—488.
( 238 )

THE UNFELT EARTHQUAKE, 230

14h. 6m." The amplitude of the recorded oscillations of the mirror
when at its maximum was equivalent to a tilting of the whole instru-
ment through 20” of arc; but in the gap, where the oscillations were

too great to be photographically recorded, it must have been greater.

In discussing these observations the first point to be noticed is
that there are three distinct phases and seven epochs, which can be
recognised in the records. The phases are (1) the first phase, con-
sisting of nearly horizontal rapid displacements of the instruments
unaccompanied by undulating movement of the ground, (2) the
second phase, during which there were distinct undulations, appar-
ently accompanied by angular tilting of the instruments, and (3) the
great oscillations commencing at about 11-45. The epochs which

can be recognised more or less distinctly on the records are—

(1) The first commencement of disturbances.

(2) The maximum of the movement of the first phase.

(3) The commencement of the undulatory movement of the
second phase, |

(4) The maximum of the same.

(5) The commencement of the larger undulations.

(6) The maximum of these larger undulations.

(7) The cessation of the disturbance.

One or all of the epochs can be determined on all the traces,
and the respective times are given in the tabular statement on

the following page, the time being minutes and seconds after
IIh. Greenwich mean time: —

! In the published account 13h. 12m. I am indebted. to Prof. R. Copeland, Astronomer
Royal, for the information that a reexamination of the record showed that the disturbance
could be traced for a longer period than was originally supposed.

( 239 )

WEN or

240 OLDHAM: GREAT EARTHQUAKE OF 1897.
Q4 vs 3 4 JA: 6
FIRST PHASE, SECOND PHASE. LARGE UNDULA-
TIONS. End of
distur-
Comm | Max. Comm Max Comm Max bance
m. S m. 8 m.s, Aum. 3s m. S. m. S
seba... 01.322.713. 1.39 a0 4 125.9 0.30975 40) ری ہدوہ‎ ien 13h
Rome, Rocca di| 17 4o 25 50 4 41 45 46 5o 13h.
Papa. A
B Ufficio 17 5 20 20 25 20 ۶ 47 o SE
centrale.
Padua e el 16 47 2270 25 45 | 28 15| 42 30 47 10 15h.
Catania . 1-17 23 de 25 55 | 26 25| 45 ° | 48 o 14h.
Siena e AG O | 19 30 20 0° 127 01 46 o 48 o
Pavia E d A9. 17 20 © ais 27 o| 45 43 48 o 12h.
Grenoble , ON Krk + 5v oe n ES ^
Potsdam e e 18 45 aoe ven ž PTT] ... eco soe
Strassburg e 48. 32 Hs un Low e M 14°5h.
Edinburgh . 18 0 eee 28 0 eee eee ... : | 14*5h.

In connection with this table and the apparent imperfection of
the records at all but the Italian observatories, the difference in type
of the instruments used must be borne in mind. دا‎ Italy, heavily
weighted pendula are favoured, whether vertical or horizontal, and the
record is made by a pen, or a pointer travelling over a moving strip
of smoked paper. This type of instrument has some disadvantages,

but it enables the recording surface to move sufficiently rapidly to

give an open and detailed record of the displacements of the
pendulum, and the heavy weight on the pendulum enables it, in
virtue of its inertia, to record horizontal displacements of the
instrument which are unaccompanied by tilting.

At Potsdam, Strassburg, and apparently Grenoble, weighted
horizontal pendula were in use, At Potsdam,the instrument was
thrown out of working order shortly after the commencement of the
shock. At Strassburg, the displacement of the pendulum was so

( 240 )

THE UNFELT EARTHQUAKE. 243

great that no impression was made on the sensitised paper, and
from Grenoble I have no details.

The bifilar pendulum is not intended to respond to arything but
a tilt of the instrument as a whole, but it is evident, from the time of
commencement of the preliminary tremors, that the movement of the
first phase was sufficiently great to affect the instrument, while the
commencement of the larger displacements indicates the commence-
ment of the second phase, The time of maximum movement cannot
be decided, as at that period the displacements were beyond the
range of recording power of the instrument,

Turning to the magnetic instruments, I have been able to collect
the following information.

Bombay.—A description of the effect of the earthquake on the
traces of the self-recording instruments at the Colaba Observatory
has been given in Chapter XI. The main results may be epito-
mised here and the times converted into G. M. T.:—

Declination, commencement 11h. 12'75m. end rıh, 41:75m. the-

trace is lost between 11h. 14°75m. and 11h, 16'75m.

Horizontal force, commencement of disturbance 11h. 11'25m,
end 11h. 44°75m. trace lost between 11h. 14:25m. and 11h.
28:25m.

Vertical force, commencement disturbance 11h. 15'25m. end:
11h, 22:75m. maximum at 11h, 16'75m. and a second minor
one at 11h, 20°75m.

Barograph, commencement of disturbance; 11h. 14'25m. end
11h. 21'25m. maximum at 11h, 18:25 h.

Batavia.—At Batavia the magnetic and electric instruments were
affected by the shock. The times taken from the original records,
for which I am indebted to Dr. S. Figee, are as follows :—

Declination, Shocks, almost imperceptible, at 11h, 22m,, G. M.
T., and others afterwards until 11h, 32m.
R ( 241 )

242 OLDHAM: GREAT EARTHQUAKE OF 1897,

Horizontal Force, First, small, shock at 11h. 24m., G. M. Th
last at 12h. 55m,

Vertical force, First shock at 11h. 23m., others until 11h, 43m.

Electrometer, (Mascart’s type) commencement 11h. 17m., end
at about 11h. 53m,

The disturbances are of the same type as those of the Bombay
instruments, but less extensive,

Potsdam —The instruments are similar to the Kew instruments
but smaller in size. All three were affected by the earthquake, The
times of commencement were— `

Declination : : :۱8ے‎ 00 T:

Vertical force . e 1185882580٤ 7

The exact time of commencement cannot be determined on the
trace of the horizontal force instrument, but there is a perceptible
. widening of the line commencing at about 11h. 10m., though the first
definite disturbance is not for some 6 or 7 minutes later. The ter-
mination of the disturbance is at about 12h. 4om. G. M, T. on the
trace of the declination instrument, andabout 12h, 28m. on that of the
vertical force instrument. On the horizontal force instrument there
are a number of small oscillations lasting till about 13h, 42m. G. M, T.

The times of commencement were measured at Potsdam on the
original records and are accurate within a limit of ‘5m. The
others are taken from photographs of the traces, for which I am in-
debted to Dr. Eschenhagen, and cannot be trusted within, say, 2
minutes of time. I am also indebted to Dr. Eschenhagen for the
informationthat the maximum displacement recorded by the horizontal
force instrument was at 11h. 48:75m. G. M. T., but, as the time
for changing the recording papers fell just as the instruments were
being affected by the earthquake, there may have BE a greater
movement subsequent to this,

Wilhelmshaven,—The magnetic instruments at Wilhelmahaven
( 242 ) |

THE UNFELT EARTHQUAKE. 243

were all affected by the shock, and I am indebted to Dr. C. Bórgen
for the following particulars, obtained by measurement from the
original traces. The tithes are reduced to Greenwich Mean Time

and correct to about half a minute سے‎

Derlination E
11h. 19'5m. first indication.
26m. first distinctly visible wave followed by several small waves.
44'5m. first strong wave.
50'5m. maximum.
560 m.
to many shocks, succeeding each other quickly,
12h. 0'5m,
13h. 28m. cessation of disturbance,
Horizontal force.—
11h. 19m. first indication.
42°5 strong disturbance ; after this the record ceases till ith 57'5m.,
after which it is again visible.
12h. gom. cessation.
Vertical force.—
11h, 26'5m. commencement.
11h. 290m, small wave.
1th. 31m. larger wave.
11h. 40m,
to a violent movement of the magnet causing a blurred,trace.

12h. Om.

Copenhagen.—The magnetographs in the observatory of the
Meteorological Institute at Copenhagen were affected by the shock.
| am indebted to Dr. A. Paulsen, the Director of the Institute, for
the following particulars, obtained by direct measurement on the
original records :—

Declination, commencement of disturbance 11h. 25'5m., end 11h.
56'5m.; maximum 11h. 36m, ; maximum angular displacement
0'75m,

Horizontal force, commencement of disturbance 11h, 16'5m. end
11h. 515m, maximum displacement o'6m. at 11h, 36m,

Vertical force.—4A disturbance of the trace was noticed at 11h,
47m,, the angular displacement is too small to be measured.

R 2 (343 )

الس عق مہ نا

2i

214 OLDHAM: GREAT EARTHQUAKE OF 1897.

Pawlowsk (St. Petersburg).—A ccording to Dr. Agamennone! the
horizontal force instrument at this observatory was disturbed at 11h.
17m. the disturbance reaching its maximuni at 11h. 22m., and ending
at 11h, 25m., but the trace is again disturbed from 11h. 39m., to rıh.
42m. though in a less degree. The declination instrument trace was
affected at the same time, but not so distinctly.

Paris, Parc, St. Maur.--According to the Bulletin du Bureau
central there was a slight disturbance of the curves of the horizontal
force and declination instruments at 11h. 27m. 40s.? |

Utrecht.—The disturbance of the declination instrument com-
menced at 11h. 18°5m., and ended at 12h. 38m. |

The horizontal force instrument was not disturbed till about 11h..
56:5m., the disturbance lasted till 15h. 47m.

Tabulating the times given, for convenience of comparison, we
have the following :—

HORIZONTAL FORCE. DECLINATION. VERTICAL FORCE,

Comm.| Max. | End. Comm. Max. End. | Comm. | Max. | End.

—Á—M——— j———— | ———— ————ÓFdB————— | ——— | —————| ———————
————

Bombay .| II 1 ? 1i AS Ir mi I1 16 B6 42 | IY. 15| 11 17 | 11 23

Batavia -| 11. 24 2 11 55 | Ep. 22 5 2327 23 ? 11 43

Potsdam . 2 11- 49 1.13942 | AFL 10 ? 12 40 | 11 18 ? I2 28

Wilhelms-| 11 19 ? 12 40 | I1 19'5| 11 50°5 = 28 | 11 26°5 ? H
haven.

Utrecht .{ 11 305 aoe 13 47 | 11185 E 12 38 s sme

Copenhagen | 1116°5 | 11 36 | I1 51'5 | II 25°5| 11 36 | 65 GA Il 47

Paris . 11 27'5 e. SS

Pawlowsk .| 11 17 | 11 22] 11 42 m

From this table it is easily seen that the magnetographs cannot
be depended on for accurate or concordant times of arrival of the

1 Boll. Soc, Sismol. Ital, 111, pt. ii, p. 290 (1897).
2 Quoted in Cosmos, XLVI, 652 (1897).

( 244 )

THE UNFELT EARTHQUAKE, 238
disturbance, or of its different phases. There is a tolerably close
agreement between the time of the first disturbance of the horizontal
force instrument and the times recorded in Italy, as those of
the arrival of the waves of the first phase, and when we consider that
this instrument is by its construction especially sensitive to
mechanical, as well as magnetic, disturbances this agreement is not
to be wondered at.

When dealing with the records of the Bombay magnetic observa,
tory the question of whether the disturbances were mechanical or
magnetic was discussed, and it was shown that though they might in
part be explained by the purely mechanical effect of the earthquake
yet they could not be wholly so explained, and must be attributed in
part at least to locally induced magnetic disturbances. An examina+
tion of the irregularities and discrepancies among the records of the
European observatories points to a similar conclusion, and it seems
as if here the mechanical effect of the earth waves was completely
masked by the induced magnetic disturbances; a conclusion which is
borne out by the limited number of observatories at which any effect
was noticed, as compared with the much larger number at which
nothing was recorded, though the instruments in use were of the
same type and in some cases identical in construction,

Whatever may be the cause of the disturbances of magnetic
instruments which are due to distant earthquakes it is evident, from
the details given above, that the records are useless for comparison
with those of instruments specially constructed to record the
mechanical displacements of the earth’s surface due to distant earth,
quakes or other causes. They will consequently be omitted in the
discussion of the distant records of this earthquake.

We have, then, records which can be utilised from 11 distinct
stations, situated at from 4,300 to 4,900 miles from the centre, as
. Shown in the following tabular statement :—

( 245 )

246 OLDHAM: GREAT EARTHQUAKE OF 1897.

I 2 7
Latitude. |Longitude.| Angle. Arc. Versin.
Potsdam + è e IESSE N ts: 2 62°36’ 4,322 576
Ischia. . : ° .| 40°45’ N | 13°58’ E 63°28’ "4,382 595
Catania . e . -| 3728 N| 1328" E 63°53 4,411 509
Rome . e e , | 42°54’ N | 11°28’ E 64° 5' 45425 603
Padua . e e 14520۰۲0۷ 7007 64°20’ 43452 610
Sia. ۲۳۰۱۰۷۰۱۱4379077 4,509 625
Pavia . ‘ e "4501/8 66°22’ 4,582 645
Strassburg = . -| 48°35’ N | 7°43’E | 60032 44 648
Grenoble i : spas T0. N Ss Lo. B; 68°14! 4,711 681
Edinburgh ^. eg [5557 Ni] 3 r WE Ost | 02 732
747

Snide . - : Sod NI 210 WI 7134 4,042

سے
————

In this table the first three columns require no explanation. The
fourth column represents the angle subtended, at the centre of the
earth by the place mentioned in the first column, and an assumed
centre in N. Lat. 25°45’, E. Long. g1%0”. The fifth column is the
length of the corresponding arc in miles, calculated to a mean
radius of 3,956 statute miles; or, in other words, the distance in a
direct line, measured along the surface of the earth. Column 6 gives
the distance in a direct line measured through the earth, and
column 7 the greatest depth, in miles, of this line from the surface.
In calculating 6 and 7; the focus was assumed to be at the surface,
and they are consequently subject to an inaccuracy on this account,
which would tend to increase the value of 6 and diminish that of 7,
In other words, the true distance in a direct line would be rather less
than that tabulated, and the true depth rather greater, The error
is, however, proportionately so small that it may be neglected.

Taking these distances, we may calculate the rates of travel,
represented by the principal epochs in the records. In doing so, we

may assume that all the waves started at the sametime from the

( 246 )

THE UNFELT EARTHQUAKE. 247
centre, and as a matter of fact they certainly did so within a minute
of time.

Taking the time of arrival of the first tremors, we have the follow-
ing result :—

APPARENT RATE OF TRAVEL ALONG

SURFACE, DIRECT LINE.

Interval in

minutes.
Miles per | Kilometres | Miles per | Kilometres
min. per sec. min. per Sec.
A ا‎ e VS 12°25 366 9°68 340 8'99
Cina ود"‎ e 12'5 353 9°33 335 8:36
Rome, Rocca di Papa . 12°75 347 9'18 329 8'71
an Ufficio centrale . Loo 369 9°75 350 9'25
Padua . e . e 11°75 379 10°02 359 9'50
Siena . - ; e 120 376 9'94 356 9'41
Pasta. acti tive le 13°25 355 9°19 347 8°64
Arithmetical mean . 12°36 364 9'58 345 9°05
Potsdam . . . - 13°75 314 8°32 299 71
Strassburg . . ٠ 13°5. 340 9'00 321 8:5
Grenoble. a è è 140 336 8'90 317 8°39
Edinburgh . ۱ ^ 13'0 353 9°33 361 9°95

That is to say, these tremors travelled between India and Europe
at a rate of about 345 miles a minute, or g'o kilometres a second, if
they travelled in a straight line from point to point, and at a greater
rate if they followed a curved line.

Turning to the time of commencement of the second phase we
have the following results :—

( 247 )

248 OLDHAM: GREAT EARTHQUAKE OF 1897.

APPARENT RATE OF TRAVEL ALONG

SURFACE, DIRECT LINE.
Interval in

minutes. | Miles per | Kilometres| Miles per | Kilometres

minute. | per second. minute. | per second.
Ischia مس‎ e e e 21'0 209 5°52 198 5'24
Catania . - 5 : 21'0 210 5'56 199 5°27
Rome, Rocca di Papa . 20'75 213 5°64 202 5°35
ss Ufficio centrale . 20°25 214 5°78 207 5°43
Padua . ~ e e 205 215 5:68 203 5:38
Siena : - > e 210 215 5°69 203 5:38
Arithmetical mean . 20°7 | 213 5°64 202 5'35
Edinburgh 5 : ` 230 213 5:63 200 5'29

Here we have a rate of propagation of about 200 miles a minute,
or 5'3 kilometres a second, lower than that of the preliminary
vibration, but still much greater than that of the sensible earthquake
in the region where it could be recognised without instrumental aid.

There remains only the group of large undulations, lasting for
about ro minutes. It is only in the Italian records that these can be
recognised, as the oscillations at this period were too extensive to
allow an impression being made on the sensitive paper where photo-
graphic registration was employed. |

For the commencement of these oscillations, we have the follow-
ing result :—

APPARENT RATE OF TRAVEL ALONG

SURFACE, DiRECT LINE.
Interval
in |
minutes. Miles per | Kilometres| Miles per | Kilometres
minute. per second. | minute. per second.
Ischia . ` - $ 375 117 111 2°04
Catania . e e E 40 IIO 105 orga
Rome, Rocca di Papa . 36°75 120 114 3922
جو‎ Ufficiocentrale i: 380 116 110 2°92

( 248 )

THE UNFELT EARTHQUAKE. 249

| APPARENT RATE OF TRAVEL ALONG

Interval SURFACE. | DIRECT LINE,
in ضضض‎ EE
EE Miles p/z | Kilometres | Miles per | Kilometres
minute. per second. | minute. per second.
Padua . - 5 : 375 119 3'14 113 2°98
Siena œ . : : 41 108 2:84 104 2°75
Pana : 2 ` 40°75 112 2°97 106 2°81

Arithmetical mean . 38°8 115 303 | 109 2°88

The important and striking fact in this table is that we obtain
a rate of propagation which is almost identical with that deduced in
Chapter III asthe probable rate of travel of the sensible earthquake.

If instead of the time of commencement we take that of maximum
movement, and take, as we may well do, the origin of these
cases to be one minute later than the earlier ones, we get very
slightly a slower rate of propogation as shown in the following
table :—

سس سے سےمس _ __Ñ—‏ ہج ی م م ےت
APPARENT RATE OF TRAVEL ALONG‏

SURFACE. DIRECT LINE.
Interval in

minutes. Miles per | Kilometres | Miles per | Kilometres

minute. | per second. | minute, per second,

APTA بس‎ RM CT ا سوج‎
Ischia . e . ۰ 41'0 107 2°83 lol 2°69
Catania . : 4 : 42'0 . 105 2:78 100 2:64
Rome, Rocca di Papa . 40°75 109 2°88 103 272
s» Ufficio centrale ٠ 40 108 2°85 102 2°71
Padua œ ` : : 41'25 108 2'86 102 2°71
Siena ' . e e S 42°0 107 2°84 102 2°69
Pavia cs. o e 40 109 2°89 103 2°73
Arithmetical mean . 4155 107 2:84 102 2'70

Before considering the meaning of these results, a short digression

( 249 )

29 OLDHAM: GREAT EARTHQUAKE Of 1897.

concerning the nature of the movements which would be propa:
gated from the earthquake centre will be necessary. They are :—

(1) Waves of elastic compression, or condensational waves.

(2) Waves of elastic distortion or distortional waves.

(3) Undulatory waves, whether purely gravitational, elastic or
elastic-gravitational, set up at the surface in and near the
epicentral region and propagated along the surface.

The existence of waves of the first two classes has long been
recognised ; they are propagated through the body of the earth and it
is to them that all earthquake effects are commonly ascribed, as was
done by Prof. Grablowitz in his discussion of the records in Ischia.
The existence of undulations of the third class, first suggested by
Professor Milne in 1888,? has not yet been fully accepted.

The rate at which these waves travel is governed by the density
of the medium and its elasticity. The rate of travel of the wave of
elastic comparison is always greatest, that of the distortional waves
less, while the only form of elastic surface undulation which has yet
been investigated has, according to Lord Rayleigh, a still slower
rate of transmission. |

Now, as pointed out by Dr. A. Schmidt of Stuttgart in 1888,? not
only it is inconceivable that the elasticity of the materials of which
the earth is composed should be unaffected by the very different
conditions of pressure and temperature existing in the deeper-seated
as compared with the superficial portions; but it is also not in
accordance with observed facts, for were such the case, the rate of
travel of the waves would inevitably be the same for tk. same rock
and the same region in the case of every earthquake,—a conclusion
which is in utter want of accordance with observed facts. We must,
accordingly, allow that the depth below the surface at which an
elastic wave is propagated has an important influence on its rate of

1 See p. 232.

? Trans. Seismol, Soc. Japan, XII, p. 112.

3 Wellenbewegurng und Erdbeben : Jahresheft, Ver. f. Vaterland. Naturkunde in Wurt,
temberg, 1888, pp. 248—270 ; noticed in Natare; LII, 631—633 (1895).

( 250 )

THE UNFELT EARTHQUAKE. 251

travel, whether this be in the direction of diminishing or, as
observations show, of increasing the speed.

The study, by means of special instruments, of the earth move-
ments due to earthquakes ata distance from their place of origin,
shows that the preliminary tremors begin before the greater undu-
lations, and that the period through which they last is longer as the
place of observation is more remote from the centre. It is found too
that the time by which the tremors precede the greater undulations is
greater than that which is due to the difference between the distance
in astraight line and along the surface of the earth ; in other words,
the rate of propagation of the tremors, which travel through the
earth, is more rapid the greater the depth of the wave path from the
surface,

In this the records of the great earthquake present no peculiari-
ties, but the marked three-fold character of the disturbance is peculiar,
The only published instance of a similar record, with which
I am acquainted, is that of the Japanese earthquake of 1894,* and the
meaning of these three phases may now be investigated,

The first phase consisted of comparatively rapid waves, whose
period is estimated by Dr. Agamennone? as not more than about ‘5s,
to ‘8s. Prof. Grablowitz' investigation of the records of his instru-
ments has already been referred to, and his conclusion that the
motion in this phase was mainly of a to and fro nature, unaccom-
panied by any considerable tilting of the ground. He expressly
refers to the wave motion in this phase as condensational, and the
most reasonable explanation is that the commencement of the dis-
turbance was due to the arrival of waves of elastic compression. If
these travelled in a straight line from the centre, the angle of
emergence would be about 30°, if the wave path wasa curved one, con-
vex to the centre of the earth, it would be higher, That is to say, the

1 E. von. Rebeur Paschwitz: Europäische Beobachtungen des grossen japanischen Erd-
bebens vom. 22 Marz 1894, und des venezolanischen Erdbebens vom. 28th April 1894, nebst
Untersuchungen úber die Fortplanzungsgeschwindigkeit dieser Erdbeben ;
Mittheilungen, 1895, pp. 13-21 and 39-42.

2 Boll, Soc. Sismol., Ital. IV, pt. i, p. 48 (1898).

Peterman.

252 OLDHAM : GREAT EARTHQUAKE OF 1897.

ground was moved bodily backwards and forwards in a direction
inclined more than 30° to the horizon.

A movement of this kind need not be accompanied by any
marked surface tilting, and any way the tilts would be too rapid for
the slowly moving pendula to respond largely to them ; consequently
the effect of the backward and forward movement on the pendula
would be much greater than that of the up and down movement, as
deduced by Prof. Grablowitz from a comparison of the records of
his instruments,

In the second phase the wave period was longer than in the first,
and there are indications of tilting of the surface of the ground.
This phase appears to be due to the waves of elastic distortion, in
which the movement of the wave particle is at right angles to the
direction of travel of the wave. As demanded by theory these waves
travelled at a slower rate than the waves of elastic compression.

An important point to note is that the rate of propagation of the
wave motion of the second phase, 5'3 kilometres per second, is in
close agreement with what is regarded as the normal rate of
propagation of the preliminary tremors, corresponding to the
distance between Italy and the epicentre. That cf the first phase,
و‎ km., per second is, onthe other hand very much in excess. Refer-
ence has already been made to the fact that the only published
instance of a three-fold division of the record, similar to that of this
earthquake, appears to be the Japanese earthquake of 1894, but too
much importance-must not be atached to this. From the nature of
the case, these waves would only be recorded, or at any rate only
separable, in the records of heavy pendula of the Italian type, and
would ordinarily be unrecorded by Darwin’s bifilar pendulum, or if
recorded, not distinctly separable from those of the second phase on
account of the slowness of travel of the recording surface and conse-
quent compression of the record. It is therefore possible that the
waves of elastic compression have been more frequenty recorded, and
that records showing a three-phase character of the trace are com-

( 252)

THE UNFELT EARTHQUAKE. 253

moner than the published references suggest ; on the other hand, the
fact that the rate of transmission of the second phase, in the case of
this earthquake, accords with what, according to the text-books, should
be regarded as a normal rate of transmission of the preliminary tremors,!
seems to show that the more rapid rate of transit shown by the
disturbance of the first phase is exceptional, or at least unusual.

This is not the place to go further into the subject. To treat it
satisfactorily the original records of distant earthquakes preserved in
the observatories of Europe require a careful reexamination, but if
the explanation of the two phases of preliminary tremors offered is
accepted, it follows that the waves of elastic compression die
out sooner than those of elastic distortion, and are only transmitted
to great distances in the case of exceptionally severe earthquakes;
in ordinary cases they have died out while the waves of elastic
distortion remain and are recorded as the preliminary tremors.
From this it is obvious that in making any deduction from the
duration of these preliminary tremors, it is necessary to be certain
which phase they begin with, whether that of the condensational or
the distortional waves.

` In the case of both these first two phases of the records we have a
rate ‘of travel much greater than that of the earthquake which was
felt, an increase which must be mainly attributed to their having
travelled through the earth at depths where pressure and tempera.
ture produced profound modifications in the elasticity of the rocks
they travelled through.

If we now turn to the large undulations which are estimated by
Dr. Agamennone? to have had a complete period of 22 seconds, a
length of 34 miles, and a rise and fall of 20 inches, we get the

1 Prof. Milne gives the following as the average rate of transmission of the preliminary
tremors for varying distances (Seismology, 8° London, 1898, p. 113).
Distance from origin 20° rate of travel 2 to 3 km. per sec.
m nrc Se m) A » s» وو‎

o
m 3,5 an 80 3 8 un am وو‎
an 100 IO وو وو وو‎

35» ر7 ود‎
* Boll, Soc, Sismol, Ital. 1V, pt. وا‎ 49 (1898).
( 253 )

254 OLDHAM x GREAT EARTHQUAKE OF 1897.

result that the rate of travel, as between the centre and Italy, was
about 113 miles per minute, or 2:98 kilometres per second, which is
practically the same as the apparent rate of travel of the sensible
shock along the surface between Calcutta and Bombay. There is
only one conclusion possible from this, that these undulations, which
are quite characteristic and distinct on the records, were not due
to waves of. elastic distortion, as has been assumed by some:
observers, and did not travel through the substance of the earth, but
along its surface. Had they travelled throughthe earth, instead of
along its surface, it is impossible to conceive that their rate of
travel would not have been affected by the depth, as was the case
with the two other kinds of wave. `

We have here a very strong confirmation of Professor Milne’s
suggestion that an earthquake sets up a wave motion at the surface
which travels outwards along the surface at a practically uniform
speed, the individual waves becoming longer and slower in the
movement of the wave particle, just as the ripples caused by a stone
falling into a pond become longer and less pronounced as they
travel outwards from the centre. |

One thing only is needed to complete the demonstration, and
that is to find the undulations which came the long way round the
earth, as well as those which came direct. If we consider what
happens to the surface undulations, we will see them spreading out-
wards in continually widening circles till they pass round the earth
and converge on the antipodes of the epicentre. Here we would:
expect to find an instrument, had there been oneset up, giving a
very confused record, but after the waves cross each other they
diverge once more, in widening circles as they travel on their way,
and in due time we should find the trace of the undulations which
have travelled past the antipodes of the centre andare on their
return journey round the other side of the world.

A simple calculation will show that if the rate of travel of these
surface undulations remained constant, the return wave should be
expected about 10 minutes past 2 P.M, by Greenwich time.

(254)

THE UNFELT EARTHQUAKE. 258

At Edinburgh, as has already been noticed, 1 am indebted to the
Astronomer Royal for the information that there is a small but
distinct disturbance of the trace at 14h. 6m.

From the observatory at the Rocca de Papa, Dr. Cancani most
obligingly sent a tracing, nearly 8 ft. long, of the record of the
horizontal pendula at that observatory, and on this I notice at 14h.
4m., two distinct though minute undulations and again between
14h. 7m, and 14h. 8m. some more. Though small these undulations
are distinct, and like the great ones of the principal disturbance,
have a complete period of about 20 seconds.

A photograph of the record at Shide shows a decided enlarge-
ment of the trace from 14h. 5m., to 14h. 7m., which is doubtless
due to the same disturbance.

Besides these I find in the detailed account published in the
Bulletin of the Italian Seismological Society, the following :—

At Leghoru oscillations of the Zromometro Egidi, are recorded
at 14h. Tom. The extent is only Imm. as against 32mm.
of the great undulations, but no other disturbance is recorded
between 13h, 42m. and 15h. 45m.

At Catania a slight disturbance of the grande sismometografo
of 25m. length of pendulum, at 14h. 12m, 41s, on the NE-SW
component?

At Ischia the horizontal pendula, which are evidently extremely
sensitive instruments, were repeatedly disturbed between 2h. and
3h. p.m. and it is impossible, without access to the original records,
to identify the undulations referred to here.

Tabulating these records, as is done below, we obtain a rate of
travel practically identical with that of the direct passage of the
great undulations from the centre. `

1 Boll, Soc. Sismol, Ital. III, pt. ii, p. 251.
2 Loc. cit. pp. 283, 284.

° (255)

256 OLDHAM : GREAT EARTHQUAKE OF 1897.

RATE OF TRAVEL,

Miles Kilomitres

Interval. Distance
وت‎ miles, Gier | debo
Edinburgh . - 5 e ` 3 1 19,968 110 7102393
Shide . . e > . ` S 23 0 20,233 112 2°96
Leghorn e e e ~ . . 3 5 20,408 110 2°93
Rocca di Papa . e . e : B. O59 20,434 114 3'02
Ditto . e e . > . 3 2 20,434 113 2°98.
Catania : a : : : : 335 8 20,448 109 2'88
Arithmetical mean . - e e 3 2 m d III 2۳

From this table we see that the time of arrival and the rate of
travel of these undulations is so close to what should be expected
if they were surface undulations travelling round the world, that we
may consider their character as such to be established, as well as is
possible in the absence of observations from intermediate stations.

The recognition of surface undulations which have travelled
round the world, may be regarded as one of the most interest-
ing results of this investigation. It is the first occasion, since
suitable instruments have been set up, on which the surface
waves of an earthquake have been of sufficient size to maintain
their character, and leave a recognisable record, after traversing five-
sixths of the circumference of the globe, and the announcement
of this may well bring the report of the great earthquake of

1897 to its close. |

( 256 )

APPENDIX A: T. D. LATOUCHE. 257

Appendix A. ^

Reports of officers of the Geological Survey of India despatched
to investigate the effects of the earthquake.

This appendix contains, with unimportant omissions, all those
parts of the reports, referred to on p. 2, which have not been in-
corporated in the body of the text. They were drawn up under specific
instructions to report only the facts observed, and to refrain from
any expression of opinion as to the conclusions to be drawn, as this
could only be profitably done after a review of the whole of the facts,
of which only part could become known to each individually.

r. Report by Mr. T. D. LaToucae, Superintendent, Geological Survey of India.
CALCUTTA, ASSAM, ۹۷۰

Calcutta. —The monument consists of a massive oblong pedestal built of brick,
measuring 16 feet 2 inches from east to west, and 13 feet from north to south,
"Uca I a A feet 3 inches in height, surmounted by a tapering
William Jones, Old obelisk, also of brick, the original height of which from
Cemetery, Park Street, the ground was 5o feet. The upper 6 feet or so of
امت‎ aide, this has been broken off, and the fragments, including a
conical stone cap which formed the summit of the obelisk, are now lying at the
foot of the pedestal on its south-west side. The longer sides of the monument
run E 20° and W 20° N.

The brickwork in falling from the top has struck and broken off the edge of
the cornice running round the top of the pedestal over the south-west side, rather
to the west of the centre, the mean direction of the break, measured from the
centre of the obelisk, being S 20° W. The stone cap struck the cornice still
further to the west, and made a separate fracture, the direction of this from the
centre being S 53° W. As the stone cap was, however, fixed to the monu-
ment by an iron pin, it is likely that the latter, striking against the sides of the
monument, caused the cap to be deflected in its descent, and the fracture caused
by the brickwork probably represents more nearly the actual direction of the
shock.

This monument consists of a square brick pedestal about six feet in height,

M with an ornamental cornice and frieze running round the

onument to rs. SA e . 1 ٠
Mary Ann Wiltshire, Old top, which is conical. This was surmounted by a brick
ee Park Street, and plaster urn, fixed to the pedestal by an iron pin.

? The urn has been shattered into fragments and fell to
the south-west, striking the cornices in three places in a direct line, the direction

5 | (257)

28 OLDHAM: GREAT EARTHQUAKE OF 1897.

of which is S 27° W from the iron pin, which still remains fixed in the top of the
pedestal,

The iron framework forming the top of the spire was thrown off, and a
number of bricks were dislodged from the portion of
the spire immediately beneath it. Some of these fell to
the north side of the tower, crashing through the
corrugated iron roof, but the larger portion fell to the southern side, and,
penetrating the roof, struck the tiled ceiling of the Church and dislodged some
plaster from the inside. A large number of bricks also fell down the sides of the
spire on all sides and lodged on the gallery ranning round the base of it. The
iron framework was projected ina direction S 313° W from the centre of the
tower, to a mean distance, horizontally, of about 4o feet, and struck the corrugated
iron roof, but only partially broke through it. The top is described by eye-wit-
nesses as having oscillated for some time, a rain of bricks falling at the same
time from the base of it, and then pitched over, coming down head foremost.
The iron stays fixing it to the brickwork of the spire have been broken through.

Cracks have appeared at each of the angles where the body of the Church
joins the tower, but the crowns of the arches supporting the tower are intact.

At the south-east corner a large portion of the cornice, which projected four feet
from the top of the wall, has fallen, as well as the heavy
balustrade above it at the edge of the roof. The corner
pillar of the balustrade, a large mass of brick masonry, has fallen almost
intact, and is now lying 21 feet from the base of the building, in a direct line
with the eastern wall. The height from the ground to the centre of the pillar,
as it stood originally, was approximately 54 feet. In falling it is possible that
the pillar struck the cornice and was thus projected further from the building
than would otherwise have been the case, but it is quite likely that the cornice -
fell as soon as the balustrade, and did not affect the fall of the latter. The
direction of projection is S 20° W. A large portion of the balustrade ad-
joining this pillar on the southern side of the roof was thrown down with it,
and is lying at the same distance from, and parallel to, the foot of the wall.

St, Paul’s Cathedral,
Calcutta.

_ Town Hall, Calcutta.

1. Brahmaputra above Goalundo. Proceeding up the river from Goalundo, I
first noticed fissuring of the banks about Sirajganj, the first place ‘the steamer
bd . stopped at. From about this point the banks are fissured
"anre ófriveribanks. on either side fe a greater or less distance from the edge
of the bank, usually about 20 or 30 yards, to the neighbourhood of Mangaldai in
Assam, a distance of some 260 miles, and fissures extend of course along the banks
of all the minor branches of the river and its tributaries within this area. As a
rule, the fissures run parallel to the bank of the river, and where this is not the
case, some peculiarity in the contour of the ground, a drop for instance from a
higher to a lower level, can usually be found to account for the change in direction.
2. At Rowmari, for instance, besides the fissures parallel to the bank of the river,

; which here runs nearly north-east and south-west, a large

i eee fissure runs to the south-east at right angles to the river

bank for a distance of at least 500 yards, when it becomes lost in a jheel. (It is

( 258

APPENDIX A: T. D. LAi OUCHE. 259

said to run to a distance of nine miles from the river, and very likely extends much
further than I traced it.) This fissure runsalong the edge of a tract of ground, on
which the village stands, rather higher than the level of the river bank, probably
marking the line of an old river channel. Sand and mud have been ejected
from the fissure to a depth of at least four feet. Other fissures branch off from this
through the higher ground to the north, one of them passing beneath the huts of
the villages. Subsequent to the ejection of the sand, the surface sank down to a
depth proportional to the amount of material ejected, and several crater-like
hollows were formed as the water drained back into the fissure (Plates X and
XI) Where the principal fissure crosses those parallel to the river bank, the
surface of the ground is broken up into a number of square blocks.

3. At Jatrapur the narrow gauge railway from Kaunia to Dhubri joins the
river. The railway bund’ is much fissured, the fissures
running parallel to the bank of a jheel along which the
railway is carried. In places the bund has subsided vertically by settlement and
spreading out on either side of the earth composing it, and the rails are bent and
twisted.

4. Dhubri. No accurate record of the time was taken at the Telegraph Office,

as the pendulum of the clock was broken by the shock. It is estimated to have
occurred at about 5-20 P.M. local time, which is 39
minutes in advance of Madras time.”
5. The gateway faces south-east. The pillars are of square section 1 foot
8 inches square by 4 feet 10 inches high, built of brick.
That on the north-east side has fallen towards N 40° E
and struck the bank alongside the gateway, rolling off into the ditch. The capping
has turned round in its fall and is lying with its point facing the pillar. The
other post is standing, but cracked through at the base.

6. This is a strongly brick-built structure, measuring outside 80 feet by 41 feet
4 inches, standing nearly cardinal, the longer side facing
N 5° E. It is divided inthe centre of its length by a
passage eight feet wide, on either side of which are two large rooms.? The flat roof
is supported as usual by strong beams, on which small battens are laid, which in
turn support the tiles of the ceiling, Inthe passage way the beams run east and
west, the battens north and south, while in the rooms the position of the beams and
battens is the reverse of this. In the passage way the battens have been dislodged
some of them having fallen out and all having moved more or less out of their
places. In the side rooms, on the contrary, the battens have not moved at all, but
the beams can be seen to have moved backwards and forwards on the walls. The
upper part of each corner of the building has fallen, in the most symmetrical
manner, for about eight or ten feet from the roof. "

7. This is a hexagonal brick structure, two storied, measuring 9 feet 6 inches
on each side (outside measurement). The door faces
S 15 E. From the top of the door-way, 7 feet 6 inches

Jatrapur.

Time of shock.

Gate pillar at Telegraph
Office.

Treasury.

Observatory.

! Embankment.
? The Cutcherry and Treasury clocks both stopped at 54 minutes past 5 P.M.
3 te, four rooms in all.

S2 ( 259)

260 OLDHAM: GREAT EARTHQUAKE OF 1897.

above the plinth, a crack runs diagonally up to the roof on the east side, passing
through the small hexagonal window on that side. A portion of the cornice has
fallen from the east-south-east and north-east sides.

7(a). The thermometer shed isin front of the observatory building. A plan

CO ME of this is given in fig. 24. It consists of a thatched roof
° up ported by eight brick pillars, each of which is cracked
through near the base and twisted in the direction of the hands of a watch

from 5 to 12 degrees.

n 10° V.

Fig. 24. Plan of pillars of thermometer shed at Dhubri.

The anemometer, which was fixed tothe top of a wooden structure 14 feet
high, was thrown down, but had been removed at the time of my visit.
8. Of these two gate pillars one has fallen due north, having broken off at
Gate pillarsat Zamin- 20 inches from ground level. The other pillar has
dar’s house. not fallen, but isleaning over at an angle of 5° from
the perpendicular towards S 15° E,
g. This was an octagonal structure, well built of brick masonry, with alternate

1 walls about 14 feet high supporting an ornamental
ا‎ na roof. Between these on three sides were low walls
supporting an iron tank above the well, the fourth side on the west being left
open to allow access to the well. Of the four walls supporting the roof three, vtz.,
those on the south-west, north-west and north-east sides, have. each fallen flat

( 260 )

APPENDIX A: T. D. LATOUCHE, 261

¿Und away from the well, while the fourth on the south-east sideis
on the E standing, but cracked through. The well was originally about 30 feet
still Murs d about 20 feet of water. It is now filled up with sand to negl
deep an below the surface of the water.

GE he greater part of the station of Dhubri is built on a low eminence of
10. The 8 gneiss, bounded on the north, east and south by the
Fissures along river, and to the west by alluvium, on which the
bank, etc. puilt. This portion of the station has suffered severely from fissuring,
bazar pF near the banks of the river. Several high bunds which run due north
especial vp to the road connecting Dhubri with Kuch Bihar have been fissured
from the 1y at intervals of a few feet, and settled down considerably.
aiii اوت‎ a letter written after the earthquake by the Deputy Commissioner,
Se pe earthquake Mr. Hallifax, to the Commissioner, Assam Valley Dis-
ege tricts, dated 24th June 1897, I obtained the following

particulars :—

At Gauripur, 53 miles to north of Dhubri, pucca buildings and bridges were all
destroyed and low-lying places filled with sand. ‘The time is given as 4-30 P.M,

At Sukhchar, a khal, 300 x 78 x 5 feet, was entirely filled with sand ejected from
fissures, which is said to have spouted up to a height of 1o cubits.

At Mankachar (Lat. 25° 32, Long. 89° 55'), the thana is said to have sunk
The river, which was 30 feet deep before the earthquake, was choked

to within 8 feet from the surface, and heavy floods occurred in

river

4 cubits.
with sand

consequence. 4
At Bijni (Lat. 26° 30’, Long. go° 45'), sand and water were forced up to a height

of 4 or 5 cubits. The time of shock is given as 4-55 P. M.

Fig. 25. Overthrow of plant-tubs at the Assistant Commissioner’s bungalow, Goalpara,

12. Goalpara, The Assistant Commissioner’s bungalow is situated on the crest
of a ridge of gneiss overlooking the river, running from south-east to north-west..

(261 )

22 OLDHAM: GREAT EARTHQUAKE OF 1897.

The bungalow itself has entirely fallen, the general direction of fall being towards
Assistant — Commis. "North. In front of it is a flight of steps eight in number
sioner’s bungalow. (see fig. 25), on the topmost of which stood two tubs,
one at each side, filled with earth. Both of these have toppled over and rolled
down the steps, both falling in exactly the same direction, to N 3° W. They
seem to have struck the third step from the top in falling and then turned com -
pletely over. It is possible that the tubs were jerked forward to the edge of the
steps by the shock and then tilted over. If they were first tilted up, it would, I
found by trial, require an angle of tilt of 35° to cause the tub to slip off the
edge of the step.’ |
The cemetery is situated on the south-south-east slope of the ridge, the
average angle of slope being 15°. The monuments are
Cemetery. E mostly of the oblong box form, common some 50 years
بس‎ of Mrs Ann ago, with their longer axes lying east to west, or trans-
verse tothe slope of the hill. Mrs. Simons’ tomb has
fallen over to S 5° E without breaking, and now stands at an angle of 24° from the

vertical.

-—
a.”
POLI
-—

Pind

i

Fig. 26. Tomb of Ensign Law, Goalpara.

Ensign Law’s tomb is of the same shapegenerally, but was surmounted
by an obelisk 5 feet 8 inches high, which broke off

Tomb of Ensign Law. at its plane of junction with the top of the tomb and
was thrown to N 5° W. As it fell, it must have
1 It seems more probable to suppose that they were projected, like the tomb of Ensign
Law. —-R. D. O.
(262)

APPENDIX A: T. D. LATOUCHE. 263

turned over, since the apex is now lying at the foot of the railings surrounding
the tomb. It just touched the top of one of the spikes of the railings in falling
and bent it slightly outwards.

- A square column at the lower end of the Cemetery has fallen over to E 30° S.
This measures 3 feet 5 inches high by 1 foot 10 inches square, with a conical cap
2 feet 4 inches square by 1 foot 6 inches high.

T wo other tombs similar to that of Mrs, Simons have slipped off their founda-
tions and heeled over to sonth without breaking.

All these tombs are built of brick masonry.

I4. The Treasury was a massive brick building standing at the foot of the
hill near the river bank, and facing N 17? E. A
portion of the front of the building has fallen out, and
the west end is also destroyed; the remainder is greatly cracked.

15. A large fissure opened beneath the houses along one side of the bazar,
running roughly parallel to the river bank, in a westerly
direction. From this a large quantity of sand was
ejected, filling the interior of the houses up to the eaves. The surface of the
ground then subsided, carrying the houses with it, so that the roofs are now
resting on the sand (Plate XXII, fig. 1). A view of the fissure at the eastern end
of the bazar is given in Plate XXII, fig. 2. A well, seen at the left hand side of
the picture, was entirely filled with sand, which was ejected with such violence
that the wooden cover of the well is said to have been hurled through the air
to a distance of several yards.!

16. Tbe Telegraph Office was destroyed entirely at the beginning of the shock

Time of shock. and the time was not noted.

17. At the time of my visit slight shocks of earthquake were very frequent.
As a rule each shock was accompanied by a more or
less distinct booming sound, apparently proceeding from
the ‘south-south-east, and generally heard slightly in advance of the shock.
Frequently the sounds were heard without any shock following, and they then
resembled very closely the well known “ Barisal Guns,” though they were not
so sharp and well defined as I have heard them in the south-western portion
of the Garo Hills.

18. Gauhati. The clock atthe Telegraph Office stopped at 5-15 P. M., local time,
which is 46 minutes in advance of Madras time. The clock is said to have been
correct within One or two minutes. The pendulum was
broken by the shock.

19. At the head of the steps leading from the bungalow to the river bank,

Deputy Commissioner's two low brick pillars, supporting a wooden railing, have
bungalow. fallen to S 30° E and S 4o? E, respectively. These.
were 3 feet 6 inches high by 1 foot 7 inches by 1 íoot 1 inch on sides, the
longer sides facing south.

20. At the entrance weretwo large gate pillars, on either side of the gateway

Commissioner’s bun- facing west-south-west, 10 feet high and 2 feet 5 inches
galow gate pillars, square. One of these has broken through at 2 feet

5 See p. 104.

Treasury.

Bazars.

Subsequent shocks,

Time of shock.

(2603)

ET

264 OLDHAM: GREAT EARTHQUAKE OF 1897.

9 inches from the ground and twisted 6°, from N 24° W to N 30° W. The
other pillar is broken through at the same level but is not twisted.
21, Of the four gate pillars at the two entrances to this bungalow,’ three are
Old Dák bungalow broken off and twisted in the same direction as that
gate pillars, Pl. XXI mentioned above. The fourth pillar has fallen to
fig. 2. S 20° W. These pillars are all built of brick masonry,
22. The coping of a small gate pillar in the compound wall of this bunga-
Civil Surgeon’s bun- low was shot off and is now lying on theground at a
galow gate pillar. distance of 4 feet 4 inches from the centre of the
pillar. The direction in which it was thrown was S 35° E.
23. Both the cutcherries were wrecked, but a great portion of the roof of
Commissioners and ۈCh is standing, merely the walls having fallen. The
Deputy Commissioners Commissioner’s cutcherry is shown in Plate XX, fig. 2.
cntcherries. The roof of this was partly supported by strong wooden
posts. The Deputy Commissioner’s cutcherry (Plate XX, fig. 1) is in
the form of a cross with equal arms facing the four points of the compass.
The roof was partially supported by round brick pillars, which, though
cracked through, have not fallen. The weathercock, seen in the photograph, was
an effigy of a peacock and has tilted over to the north, so that its tail is hang-

` ing down beside the rod supporting the letter N.

N

Fig. 27. Plan of overthrown gate pillars at Gauhati,

1 This is the same building as the Telegraph signallers quarters mentioned in Chapter XIV.

(264 )

APPENDIX A: T. D. LATOUCHE. 265

24. On the west side of the Deputy Commissioner’s compound a low
wall runs along the road, on which stood a balustrade
of clay pilasters supported by pillars at intervals of
about 11 feet, Nine of these have fallen, most of them to about S 20° E, one
to N 70° W and another to N 35° W. The wall runs E 32 N, W 32° S.
One of the pillars has broken off at the top of the wall and twisted, without
falling, 10°, from N 25° W to N 15° W.

At the gateway at the east end of this wall both pillars have fallen, one
due west, the other to E 10° S (see fig. 27). At the western gateway there were
similar pillars which have broken off at the height of the top of the wall, three feet
from the ground, and fallen, one due N, the other to N 20° W.

25. Most of the tombs here are of the common square or oblong box shape
built of brick masonry and plastered. That of G. T.
Bayfield, died September 1840, was surmounted by an
obelisk, the greater portion of which has fallen to S 20° E. A similar Obelisk close
by (F. Gomes, died September 1848), has fallen to S 25° E. The monument shown
in Plate XXI, fig. 1, is of precisely similar construction, but in this case the obelisk
is broken across at about half of its height, and the upper part has twisted 48°,
from N 18? W to N 30? E. Close to this are two tombs standing on a common
plinth and precisely similar in all respects, but instead of conical obelisks they
were surmounted by cylindrical brick pillars 1 foot 6 inches in diameter and about
4 feet high. Of these one has fallen to N 45? W while the other has gone exactly
at right angles to this, viz., to S 45° W. Both pillars have been entirely broken up
by the fall, but in the first case the base of the pillar is lying at a greater distance
from the base of the tomb than the rest of the pillar. The tomb measures 3 feet
square by 4 feet 6 inches high, and the base of the pillar is lying at a distance of
12 feet from the corner of the tomb nearest to it. The top of another obelisk
(Samuel Fleming, died May 1853) has fallen to N 45? W. The lower part of the
tomb on which this stands is also cracked through, at a height 1 foot 7 inches
from the plinth, This tomb measures 4 feet 9 inches high by 2 feet o inches
square.

Only one of the more recent tombs has been affected by the shock (Miss Lamb,
died April 1875). This is a small marble pedestal standing on a brick plinth
and surmounted by a cylindrical marble pillar 3 feet 5 inches high by 112 inches
diameter. The pillar was thrown off to the south and fell, without striking the
low railing surrounding the tomb, and is now lying pointing east and west at a
distance of 8 feet 6 inches from the centre of the pedestal. It may have rolleg
somewhat after touching the ground.

This cemetery is situated on the alluvial ground to the south of the station.

26. The new cemetery is situated on the flanks of a low hill of gneiss to the
east of the station, and has suffered little or no damage
from the earthquake, except that the wall surrounding it
¿has fallen at the north-west and north-east corners and near the south-west
corners, A marble slab on the tomb of R. J. Eades, died 1884, is cracked across
the middle, the crack running N 10° E.

Compound wall,

Old cemetery.

New cemetery.

( 265)

266 OLDHAM: GREAT EARTHQUAKE OF 1897.

27. This is a perfectly new brick building, but is greatly damaged by the
shock, It faces S 10% E. The centre gable has fallen
۱ out on both sides of the building to south and north
respectively, the brickwork having broken up into large masses. The end walls
east and west are cracked but standing, the cracks running diagonally in both
directions, north and south, at an angle of about 45°, from the corners of the

windows in centre of walls, È
28. This Temple is an octagonal building with a domed roof, built of tiles
Old temple near 9 inches X7 inches Xi} inches in size. The sides facing
steamer ghat, E 15^ S and W 15° N have fallen outwards altogether;

and those facing north and south are partly fallen.

Railway station.

29. This is a brick structure with a flat roof supported by numerous arches
with minarets at each corner and on either side of the

gk upra LEE. at the large central arches. The larger side, towards the
" river, faces west. Only one of the minarets has fallen,

that on the north side of the central arch towards the river. This has fallen due
west. The two side arches at each side of the central arch, and those at each

end of the building, are cracked at the crown.

30. At the western end of the Gauhati Bazar is a bridge of three girders

e carrying the Grand Trunk Road over a small stream
goatee on and Zug which here joins the Brahmaputra. The original ac?

of the bridge, as measured along the hand rail, was
وو‎ feet 4 inches, while the present length, between the same points is 97 feet
10 inches, The bridge has therefore been shortened 18 inches. This has been
caused by fissuring of the banks on both sides of the stream, the abutments
having been carried forward. One of the piers has been tilted over, probably by the
thrust cf the girder. There are no cracks in the abutments.

31. On the north bank of the Brahmaputra, opposite Gauhati, are some old

. temples standing on a rocky promontory immediatel
ER E. GEE over the river. These were built of tiles similar to uc
in the temple above mentioned. All except one, which
is at the base of the promontory close to the water, have fallen, the tiles having slid
off each other in all directions. The dome of the one which is standing is split
through the middle, the crack running N 15? E and vertical.

A small structure consisting of a corrugated iron roof supported by four low
brick pillars, erected over a lingam near the temples, has fallen over bodily
towards E 10° S.

32. The accounts of the earthquake received from the surrounding country all

speak of fissures opening along the banks of rivers and

Pos of earthquake in consequent subsidence of the surface, also of quantities of
j sand being ejected, filling up the beds of rivers and

jbeels and causing floods. The Sub-Deputy Collector of the Borpeta Circle says
that, while he was proceeding by boat down thé Singur river towards Borpeta,
which lies to the west-north-west of Gauhati, he heard, when near Kahi Kuchi,
alow rumbling noise occasionally, apparently coming from the direction of the

( 266 )

APPENDIX A: T. D. LATOUCHE, 267

Garo Hills, to the south-west. The shock was immediately preceded by a noise of
the same kind. The river became filled with sand, and the water flowed over the
surrounding country. At Borpeta the water continued rising till the roth June
and the station was flooded. i |

At Patti Darrang, 10 miles to the north of Gauhati, an old£stone bridge, said
to have been built by the Mohamedans, has fallen.

At Hazu, 16 miles west-south-west {rom Gauhati, the Mahdub or Madhava
temple," said to have been the oldest in Assam, has also been destroyed.

33. Tezpur. The time at which the shock occurred was not observed in the
Telegraph Office here.

The east wall of the Church has fallen outwards. The north and south‏ ےہ

P walls are considerably cracked. The cracks run vertically
Ren. upward from the crowns of the arches over the windows.
At the western end of the Church isa small belfry supported by tall wooden posts
let into a brick-and-cement plinth, raised 2 feet 9 inches above the ground. The

posts have rocked from east to west, forcing out the bricks and cement at the edge
of the plinth opposite to each post.

35. In the Cemetery very little damage has been done and no good measure-

ments could be obtained. An inverted ghurra on the top

de of a square pillar 8 feet high, has fallen over to south, and

an Obelisk On another tomb has been cracked through at about one-third of its
height from the top, the plane of the crack dipping slightly to south-west.

A large pillar on one side of the gateway, measuring four feet square, has

fallen to pieces, the greater portion having gone to the west. The corresponding

pillar is badly cracked but standing. These were built of very loose masonry.

« 36. Beyond the cases above mentioned 1 could find no overthrown gate pillars

pu E or anything of the kind which would enable one to de-

À termine the direction of the shock at Tezpur. The banks

of the river here were not fissured, and I did not notice any fissures to the east of

Mangaldai. Numerous large fissures are, however, said to have occurred at

Nowgong.

37. Assam-Bengal Railway. I went out along this line as far as the bridge
over the Kopili river, about 41 miles from Gauhati. The rock cuttings, in gneiss,
have not been affected in the slightest degree by the shock, but where the line
passes over alluvium, the embankment has settled down carrying the rails with
it. Many of the culverts are badly cracked, apparently from the same cause as
has affected the bridge mentioned above at Gauhati, viz., the fissuring of the banks
of the streams and the consequent sliding forward of the abutments and wing walls,
The piers of the large bridge over the Kopili are cracked through horizontally at
about 2 feet above the ground level, and the girders have shifted lengthways on
top of the piers. This bridge runs east to west.

38. Shillong. The time of the shock on the 12th June was not observed accu-
rately, but it is said to have occurred at 5-15 P.M., local-time. Since the first

dod shocks occurred the signallers have amused themselves by
PR NOS telegraphing a certain signal to Gauhati or Sylhet,whenever
! This temple was built in 1672 Saka, under orders of the King Pramatta Singha,

( 267 )

268 OLDHAM: GREAT EARTHQUAKE OF 1897.

they happened'to be at the instrument and felt a shock, at the same time receiving’
a signal from either of those places, if the shock was felt there. The Telegraph
Master assures me that in all cases the shocks are felt absolutely simultaneously
at these places. 1 asked him to note the exact second at which a shock was felt
in Shillong, and to ask the signallers at Gauhati and Sylhet to do the same,
but he assured me afterwards that no difference could be detected.

He also informed me that an Assistant was sent downto Nongpoh, on the
Gauhati road, as soon as possible after the shock of the 12th June, to restore com-
munication with Gauhati. On attempting to signal through a single wire, with
return circuit through the ground, it was found that as each earthquake shock
occurred the current was interrupted, or even reversed. What is called a “ closed
circuit” was being used, that is, the current was kept continuously flowing through
the wire, and interrupted by the key only at the moment of sending a signal.
Apparently the earthquake shocks set up currents in the earth, for, when a second
wire was used instead of the earth asa return circuit, no effect of the kind was
observed.

39. A large number of the monuments in the cemetery were affected by the

۱ shock. The cemetery lies on a spur at the west end of
Cemetery. the station, the ground sloping at an average angle of
about 7° to south and south-west. The greater number of the monuments, of
which plans are given, lie close together on the more level ground on the crest

of the spur.

Major Willans, d. July
1886.

This is a marble slab, 43 inches thick, resting on a
plinth of quartzite." The slab has been jerked towards
E and at the west end, slightly to S.

G. R. Nicholls, "d. A marble slab, carved into a cross, resting on a
May 1889. quartzite plinth. The movement of this has been
towards S-E.

This is a granite cross standing on a base of the
same rock forming three steps, the lowest of which is
tilted to Eat an angle of 2° 50”, while the upper two have
fallen over with the cross and are still attached to it.

This is a similar cross but built of marble. The

Mrs. Spring, d. Decem- lowest step has tilted over at an angle of 8° 30", the next
ber 1885. has slid off and is resting against it, while the cross with
the topmost step attached has fallen to E.

This is a small marble pedestal, resting on a plinth

Mrs. Rossenrode, d. of quartzite slabs covering the grave, and supporting a
Jour small marble cross. The pedestal has twisted round 26^,

Pl, XXXVIII, fig. 2. 1 ; 0

the eastern side now facing E 26° S. Thecross has
fallen and is lying in the position shown at the foot of the pedestal.
This consists of two marble slabs, the upper one

Mrs. Walker, d. July carved into across. The lower slab has tilted over to S,
ibi at an angle of 10°, and the upper one slid oft in the same

direction. :
1 The quartzite, of which all the plinths and some of the tombs are built, is quarried from

the quartzites of the Shillong series in the neighbourhood.

( 268 )

Mrs. Maxwell, d. Feb-
ruary 1889.

APPENDIX A: T. D. ۰ 269

Mrs, Sherriff, d. August This is a similar tomb to the last, but in this case both

1881. slabs have been jerked to the north-east. The upper
EA one then slid off and is lying beside the lower.

SNE SE This is a marble cross standing on a marble pedestal,

d Maitland, d. supported by four narrow strips of marble. One of‏ ا

these strips on the southern side has been jerked

to S-W and the pedestal has slid off to S. The cross then fell over to
S 15° W, the top of it catching on the railings of the adjoining tomb.

This is a small dome-shaped marble slab supported
Douglas Gray,

d. June on a quartzite plinth. The slab has been jerked to
1879. SE

This is a freestone slab carved into a cross, resting on

Captain Cookesley, d. a thinner slab of similar material, the whole supported

April 1872. on plinth of quartzite blocks. The lower slab has

moved two inches to E, while the upper has moved three inches further in the

same direction, and has been jerked off to S, the edge of it. resting on edge of
the plinth.

A marble slab with cross carved upon it, resting ona

peces A quartzite plinth. The western end of the stone has been

jerked round to S,while the eastern end has hardly moved.

This is also a marble slab, similar to the last but

C. A. Eglinton, d. June supported on four oblong strips of marble. The slab has

Ko been jerked to N E, while the strips at the W and 5 sides
of the grave have gone in the opposite direction.

| A small marble cross three inches thick lying on a
tomb without quartzite plinth. The cross has been jerked its own width
due S.

This is a small freestone slab carved into a cross
0 Goldney, d. July resting on a quartzite plinth. The slab has been jerked
Jost» towards S E.

In this case the marble slab carries a small marble

884. cross, supported on three steps near the west end. The

Pl, XXXVIII, fig. 1. cross has not fallen, but the west end of the slab support-
ing it has been jerked to the south, being twisted through an argle of 15°.

This is alarge marble slab resting on a quartzite

Miss F. Fitzpatrick, d. plinth. It has been jerked about two inches to E, and

Seats ری‎ the eastern end has moved slightly to south. This tomb
lies on the slope of the spur near the south-east corner of the cemetery.

This is a small tomb consisting of two oblong blocks

of quartzite with a small marble cross let into the centre

of the top. The blocks have been jerked towards S rather more at the western
end than at the other,

Small
date.

R, D. Prazer, d. June

Rose Skene.

: This is a similiar tomb to the last, but in this case
Cecil Skene, : :
there are four blocks of quartzite covering the slab sup-

( 269 )

270 OLDHAM: GREAT EARTHQUAKE OF 1897.

porting the cross. The whole have been jerked in the same direction as the other,
but the block on the south has separated from the others and fallen over the
edge of the plinth.

At the top of the cemetery near the north-east corner three crosses have fallen
due W and broken up. The headstone of another grave near these has broken
off a few inches above the ground and fallen to E.

40. The/pillars on either side of the gateway are built of cubical blocks of stone
with a heavy stone coping, facing the four cardinal
points very nearly. That on the south side of the gateway
Pl. XXXIV. has fallen due N. The other is standing, but has
twisted about 5°, from N8? E to N3? E. The cubes have each an iron pin let into
the centre, which fits into a hole in the cube above.

AI. These pillars are similar to the last, but the blocks measure 16" x 16" x 10" and

the coping 21" x 21" x 3". One of them has fa!len to N 10?E-

Inglisby gate? pillars. The other has not fallen in any particular direction, but

the stones are lying scattered round the base. These

pillars were not able, however, to fall freely, as the wire fencing surrounding the
compound was attached to them. `

Two pillars on the opposite side of the road from this house, built of oblong
blocks of stone, have both fallen due S.

42. These were two pillars built of rubble stone masonry,
وا یں‎ Hall gate pil- the longer sides of each facing north and south. Both
have fallen, the one due N and the other due S.

43. These are two strongly built pillars of cut stone. neither of which has fallen.

, They are both a good deal shattered near the base, and
Lx e E gate the one on the western side of the gateway has shifted
Pl. XXXIV. bodily above the second course from the base to the

north-west, twisting slightly, from N15?E to (18۴ at
the same time. The other pillar has also twisted, the greatest amount of twist
having taken place between the 4th and sth courses from the base, from N15°E to
Nig°E. The lower seven courses of this pillar.are more or less shattered. Each
pillar supports a heavy iron gate, which was standing open at the time of the shock.

44. One of the pillars of the gateway of this house, built of squared stone blocks,

. has fallen to N 10° W, the pillar faced N 40° W, and
Col. Macgregor’shouse was 5 ft.2 in. high by 2 ft. square. The correspond-
Sate Sao: ing pillar to which the gate is attached is much shaken `

Fenton’s Hotel gate
pillars.

but has not fallen.
45. Of other pillars of which I was not able to get the dimensions, as they
Other gate pillars. have broken up entirely, the directions of fall were—
(1) .S SR.
(2) S 25° W ? stones probably moved since fall,
(3) S.
46. The Willans memorial stood in the club grounds, and consisted of a
square pedestal supported on two steps, and surmounted
Willans memorial, by an obelisk. The whole was built of blocks of Shillong `
1, quartzite. The apex of the obelisk, which was a square

(270)

APPENDIX A: T. D. LATOUCHE 271

pyramidal stone measuring about 1 foot on sides, had been carried from the position
in which it fell before I saw it, but I was able to identify four of the blocks immediately
beneath this, and the positions in which they fell are shown in the plan. The
three blocks seen in the sketch on the top of the portion of the +0 still
standing are twisted slightly towards the east.

47. The large bridge on the Gauhati Road about 13 miles from Shillong, over the

. Umkra river, has suffered severely. The abutment on the

A CR on Gavhati South-east side fell entirely, carrying the girders with it.

The two piers and the abutment on the north-west side,

which are of more recent construction, remained standing, though somewhat

cracked. It appears that the piers were recently widened and that the vertical

cracks near the lower sides of them occurred at the junction of the newer and older
masonry.

48. Numerous landslips have occurred along the steep hill sides between
Lalo; Shillong and the crossing of the Umiam river, 8 miles
A from the station. At the Bishop's Fall, about 2 miles
from Shillong, the precipitous cliff on the right of the fall, down which the path
was carried, slipped down entirely into the basin at the foot of the fall. The crest
cf the fall was not affected, as it is formed by a strong dyke of diorite. The nearly
vertical quartzites have been shaken away from this dyke. Some fine slips
are seen on the hillside facing the fall. A view of these is given in Pl. III.
49. At the Khasia Bazar at Maokhar, just outside Shillong on the Gaubati
road, was a collection of the large monoliths of quartzite
set up in former times by the Khasias as ancestral
memorials. Several of these have fallen, the majority between W and S W,
but one or twohave fallen due N. Some of them are broken through at ground
level or a foot or so above it.

Khasia monoliths.

So: Maophlang. At the Dak bungalow at Maophlang, 14 miles to the south-

west of Shillong, there were four large monoliths in a line running W. 30° S to

hata کا‎ E 30° N. Two of these have fallen, both to S 30° W.

One ofthese was 20 feet high by 4 feet 5 inches broad

and 9 inches thick at the base, slightly tapering tow ards the top. This has broken

off at 2 feet from the ground, The other was 14 feet high and has broken off
at ground level.

51. Mr. Evans, the Missionary at Maophlang, informed me that soon after
MET TT the earthquake his attention was called by one of his
level in hills, converts, a Khasia, to the aspect of the hill immediately to
the west of the village. These hills are separated from

that on which Maophlang standsby a deep valley, through which one of the
tributaries of the Bogapani runs, It appeared to them that beyond the hill on the
west side of this valley they could see more of the distant hills than before, and they
came to the conclusion that the intermediate range of hills had subsided. The
furthest peaks they can see from Maophlang are some four or five miles distant. Of
course an elevation of the more distant hills, or of that on which they were standing,
would produce the same apparent effect as a subsidence of the intermediate hills

( 271)

272 OLDHAM: GREAT EARTHQUAKE OF 1897.

and ifany change of level has taken place, I should think it more likely that it
would be elevation than subsidence. The hills to the west of Maophlang, judging
from the number of landslips visible, have evidently been very severely shaken.

52. The Khasia path from Maophlang to Cherrapunji, which is part of the
old main road across the hills from Tharia Ghat to
Gauhati, has not suffered much, as it usually avoids the
side slopes of the hills. The cart road, however, from Shillong to Cherra has
entirely disappeared in places, wherever it was carried along the edge of the
scarps above the stream, which runs ina deep gorge to the east of Cherra, and
is everywhere badly fissured.

53. Cherapunji. The monument erected by the Supreme Government to the
memory of David Scott is a very massive structure, built of large squared blocks
of sandstone, bound to each other with iron clamps. The
upper portion of the obelisk has fallen, mainly towards
the S W, but some portion has also fallen towards N N E. One of the
loosened stones is still lying on the top, overhanging the southern side by
about i of its length. The whole monument above the two steps at the base
has moved bodily towards S W. The stones of the lowest course of the pedestal
have been shaken apart from each other, the one on the west side having been
jerked out to the edge of the step below.

Road to Cherrapunji.

David Scott's monument.

Fig. 28. Plan of overturned gate pillars at entrance to Inglis’ house, Cherrapunji.

(272)

APPENDIX A: T, D. LATOUCHE. 275

54. At the entrance to the avenue leading to Mr. Inglis’ house stood two lofty

Mr. Inglis’ house gate gate pillars, which formed a conspicuous object in the land

pillars. scape. No gates were attached to them. They were

built of rubble stone masonry. Both have fallen, but in different directions, as

shown in the plan, fig. 28. The material of one lies scattered to a distance of 19
feet from the base of the pillar, and the other to 2r feet.

Close to the gateway of the house a small stream crosses the road, over which,
when the house was inhabited, a small aqueduct was carried on a series of brick
pillars. One of these on the edge of the stream has fallen, and the corresponding
one on the other side has cracked across at the base and twisted through 5° from
N 25° E to N 30° E. The pillars are built of brick.

At the gateway leading into the compound of the house an ornamental
pillar standing on the wall surrounding the compound has broken across at the
top of the wall and twisted through 5° from N 54° Eto N 58” E. The height of
the fracture from the ground is 4 feet 6 inches. The two gate pillars have fallen
in a heap in the gateway, that on the east side having fallen to south-west, These
pillars are built of rubble stone masonry.!

55. On the road from Cherra to Maosmai, there are two gate pillars, one of
which has fallen to N 50° E, while the other is not even cracked. These are
cylindrical, 2 feet in diameter and 7 feet hich, built of rubble stone masonry.

56. All the tombs are of the oblong form with sloping tops common 50 years

C ago, and are all built of rubble stone masonry. Very
emetery. 5
Plate V. few are broken up, but nearly all have sunk down into
the loose sand beneath them, and are leaning over at
various ume to the north, The cemetery is situated on the top of one of the
small knolls of sandstone which are scattered over the Cherra plateau. This
sandstone originally rested upon the limestone of the plateau, which has been
dissolved away from beneath it, and is accordingly much broken, The earthquake
seems to ‘have shaken the surface down into a perfect quicksand, into which the
- tombs sank.

57. Maosmai. At the village of Maosmai, on the edge of the cliffs about 3 miles

south of Cherra, there are two well known groups of Khasia monoliths of unknown
: À antiquity, from which the village takes its name (the
ee oes کم‎ e oath of allegiance), In each group there

are five stones ranging from about 12 to 18 or 20 feet high. They stand in a line
facing due east on the edge of the stream running through the village. 1 found that
one stone in each group had fallen, in each case due W.

In the larger group the fallen stone measures 14 ft. X 3 ft. x1 ft. In the smaller
I2 ft,X 7 They were let into the ground to between a foot and 18 inches.

58. I asked the missionaries both at Cherra and Maophlang, whether it was
likely that, in case any of these monoliths, which are found all over the hills, had
been overthrown by a former earthquake, the Khasias would have set them up
again. They said that it might be done in the case of well known and venerated
stones like those at Maosmai, but in ordinary cases, where they were put up by the
members of a family in memory of an ancestor, no one after a generation or two

1 For plans of these two pillars see plate XXXV.

T | Genen, 3

374 OLDHAM: GREAT EARTHQUAKE OF 1897.

would take any care of them. It seems then that asthe erection of the monoliths
must date back to a remote antiquity, and as the Khasias have for generation?
neglected the custom of putting them up, no shock at all comparable to that of
the 12th June last can have visited the country for very many years,

59. Sylhet. The Telegraph Master can give no accurate information as to the
time the shock occurred, but says that it was between 5-15 and 5-20 P.M., local

Time of shock. time, which is 47 minutes in advance of Madras time.!

60. All the public buildings in Sylhet suffered more or less, but many of them

are still standing. Several are built on a narrow strip of

Public buildings. land between the bank of the river and a large tank, the

Nawab Talao, and itis likely that they were affected’
quite as much by settlement of the ground as by the shaking they received. Since
the earthquake occurred many of the cracks in these buildings running parallel to
the edge of the river or tank have opened out, more than at first, which fact
certainly points to a gradual settlement of the soil.

61. This pillar stood at the head of a flight of steps leading down to the water
at the north-east corner of the Nawab Talao. The corre-
sponding pillar at the other side of the steps has not fallen
and is not even cracked. The fallen one broke off at
ground level, and apparently fell in a northerly direction, but struck the wall on
that side and rolled off to the west.

A gate pillar at the side of the road near the Government School fell towards
+۴ E and broke to pieces in the fall. Another near the Deputy Commissioner’s
bungalow fell due N.

62 The cemetery is situated on a low knoll near cantonments, and about 3 miles
GER north of the river bank. Very few of the tombs have
been damaged by the shock. They are nearly all fairly

old and built of brick masonry. ا‎

A tomb near the east wall, without any inscription. It was surmounted by a
plaster urn fastened to the top of the tomb by an iron wire. The urn has fallen
over to S 25° E and is now lying on top of the tomb.

Pillar at side of Nawab
Talao.

Fig, 29. Overturned tomb in Cemetery, Sylhet.

1 A clock in a small tower near the river fell with the tower and'stopped at 5-15 P.M,

وو

APPENDIX A: T, D. LATOUCHE, 275

A low brick pillar, fig. 29, built of bricks rather loosely put together. The
pillar broke across at about و‎ inches from the ground and fell due east.

A small brick?pillar 2 feet 7 inches high by 1 foot 2 inches in diameter to the
north of the last has fallen due south. The top has turned over in its fall and is
lying 3 feet from the centre of the pillar. A portion of the side is lying further
away in the same direction at a distance of 5 feet from the pillar.

A square column at the south side of the Cemetery has fallen due south-
west and broken up almost entirely. The pillar was probably about 4 feet
high and has broken off at 18 inches from the ground.

63. The cantonment was situated on some rising ground to the north-west of
the Civil station and nearly a mile from the river bank,

A Hindu temple near the hospital, the four sides facing
‘the cardinal points, has been cracked in a curious way as
shown in fig, 30, the cracks running diagonally from corner to corner of the
base. On the sides facing north and south a horizontal crack runs along at
the top of the base beneath the dome, On the west side there is an arched door»

Cantonment.
Hindu temple.

Fig. 3o. Hindu Temple in cantonments, Sylhet. `

way and two cracks run diagonally upwards from the crown of the arch to the
base of the dome. The wall surrounding the temple has fallen in the north and
south sides, in each case to south, but is standing on the east and west.

A gate pillar at the foot of the hill on which the hospital stands has broken off

T2 L 275 )

276 OLDHAM: GREAT EARTHQUAKE OF 1897.

at ground level and fallen to W 18? N. . It stood at the end of a culvert, the wall
of which may have interfered with its fall. The corresponding pillar on the other
side of the culvert is not cracked. Both are built of brick masonry, 1 foot 10 inches
square, The broken piece is 5 feet long.

64. At the entrance to the bungalow occupied by the Executive Engineer, on
the main road leading north and about a mile from the
Vertical boiler station, was a well from which wat mped by an
overturned. SEL 1 AUC we e y
engine with a vertical boiler. The boiler has fallen over
to W 30° S. Its dimensions are, height 7 feet 9 inches, diameter 3 feet 6 inches ;
chimney of sheet iron, height 4 feet, diameter 10 inches.

65. The Church has fallen entirely. At the gateway in the east wall were two

large pillars of brick, both of which have partially fallen.

Church gateway. The one on the north side of the gateway apparently fell

due north, struck the churchyard wall and rolled off to
east. The other fell towards W 20° S.

The churchyard was surrounded by a low wall on which stood a balustrade of
clay pilasters, supported by brick pillars at intervals. On the walls facing north
and south these have nearly all fallen to the south, but on the east and west
walls they are standing.

66. Sonamganj. Everywhere in Sylhet, so far as I saw, the banks of the rivers
are fissured for some distance from the edge, inthe same way as those on the
Fi ۱ Brahmaputra. At Sonamganj, where the river runs nearly
issures in banks of : 2 5 ,
rivers and jheels. due west, I noticed a considerable difference in the way
in which the shock had affected either bank of the river.
On the north side a large slice, some 50 or 60 yards broad, had been detached and
slipped forward and downward, the portion furthest from the river sinking
more deeply than the edge of the bank, so that there is now over 30 feet of water
over the inner edge of the slipped portion, while over the former edge of the bank
the depth is only 8 or ro feet. As the houses are built close to the edge of the bank,
this being the only ground in the country that is above the usual flood level, they
were carried down, and the inhabitants, to the number of about 40 were drowned.
On the southern bank, however, there was no subsidence of this kind, but the
ground was fissured to a distance of 30 or 40 yards from the edge of the bank by a
number of cracks running parallel to each other and to the bank. It seems that
on the north side the bank was not sufficiently supported by the water and so slid `
forward, but at the moment of the shock the water was violently driven over to the
south side of the river and supported that bank.
67. At the time of my visit to the country round Sylhet, the rivers were in high
flood, the water over large areas overflowing the banks.
ا‎ wr d n So long as they remain full it is probable that the banks
will not slip down, since they are supported by the water.
But as the rivers fall it is almost certain that the fissured portions of the banks
will slip down, at any rate until they attain their former slope. A large amount
of silt will thus be thrown into the bed of the river, far more than the water, with
its diminished velocity during the cold weather, and its very slight fall, will be

( "56 5)

APPENDIX A: H, H. HAYDEN. 237

able to carry away at once. It is extremely likely therefore that when the rivers
rise again next rainy season, if the rainfall should be normal or at all above the
normal, unprecedented floods will occur, and it is possible that considerable
changes may take place in the course of the rivers. Moreover, as a consequence
of the enormous landslips that have occurred in the hills to the north of the Surma
valley, the hill rivers are bringing down an enormous quantity of silt, and this wil]
tend to raise further still the beds of the riversin the plains. I have already called
the attention of the Chief Commissioner of Assam tothe possibility of there being
very high floods in Sylhet next year.

2. Report by Mr. H. H. HAYDEN, Assistant Superintendent, Geological
Survey of India.

CALCUTTA, NORTHERN BENGAL.

Calcutta.—T he area dealt with in the following notes includes the parts of
Calcutta lying in the neighbourhood of Theatre Road, Park Street Cemetery,
Lower Circular Road, Harrington Street, and a portion of Alipur Lane. The notes
refer almost entirely to the effects of earthquake on the houses comprised in the
above areas ; many of the places to be referred to were visited by me in company
with Mr. LaTouche; who has, no doubt, already handed in the chief observations
of scientific value.

As regards the direction of the shock as 5hown by fallen objects; observations
were taken of the direction of fall of the Cathedral spire and portions of two
tombs in Park Street old cemeteries, with a few of minor importance, chiefly
relating to the fall of portions of walls. Measurements, etc., were made by
Mr. La Touche, relating to the Cathedral spire and certainly to one of the monu-
ments in Park Street Cemetery, iz., that of Sir Wm. Jones. The other tomb was
also examined by me in company with him, but having tioted down the results
myself, I shall give them here. in case he has not already done so. The monu-
fnent is in the northern part of the Old Park Street Cemetery, and is that of Mrs.
Mary Ann Wiltshire (1822). [t is a massive cuboidal structure, built of brick
and mortar, and havirig a somewhat pyramidal upper portion; which was sur-
mounted by an urn. This urn, which was merely made of plaster built upon a
cylindrical iron rod, was thrown down to S 243? W. Another monument in
Park Street old cemetery is of interest owing to the fact that, although probably
broken previously to the earthquake, yet it did not fall owing to the shock.
Although it was carefully examined by Mr. LaTouche and myself, we could
find no indications of the fracture being a recent one; in fact, it had every
appearance of having been in the same condition for some years.

Among buildings, the worst falls in this neighbourhood took place at No. 8,
Circular Road, and No, 44, Circular Road.

At No. 8, Circular Road, much of the northern side of the House fell out to
the north, while very bad cracks also ran through the house from north to south.

The Baptist Chapel, No. 44, Circular Road, was badly damaged, and the root
of the porch fell in. The building faces E 7? N.

The porch, which, according to the Baptist Minister, was a subsequent

/ و

278 OLDHAM: GREAT EARTHQUAKE OF 1897.

addition made about the year 1850, was in reality little more than a “lean-to,” that
is to say, it was not bonded into the main portion of the chapel, but was merely built
against it, the beams only being built into the main wall. Consequently after the
shock took place, the periods of vibration of the main building and of the porch
being different, these two structures must at some particular moment have been
vibrating in opposite directions s the result of this was that while these two elements
were travelling away from one another, some of the wooden beams supporting
the roof of the porch and running nearly east to west drew out and fell down, but
when the two vibrating bodies were approaching one another the beams were
thrown into a state of compression. Some of the beams of the porch were of wood,
while a few consisted of iron girders putin at different times to replace unsafe
wooden ones. One of these iron girders furnishes ample proof of the existence
of the above strain, for it has remained fixed at both ends but has been bent
in the middle owing to horizontal thrusts from the ends.

The remaining observations are of no great value, referring chiefly to the fall
of parapets and to the directions of cracks through the houses.

‘The cracks may be divided into two sets, those running north and south and
those running east and west ; the former would be due to the fact that the structures
vibrated east to west, and the latter to vibrations at right angles to that direction :
and during the earthquake the existence of these vibrations was clearly seen.
A good example is found in No. 13, Theatre Road, which house I was
carefully watching throughout almost the whole duration of the earthquake.
When I reached the compound (a few seconds after the first tremor), |
found that the house was violently vibrating east to west, and soona crack opened
through the kouse from north to south, and the western third of the house could
be seen vibrating outwards, the large cracks opening as much as four inches and
then closing again. This was observed from the southern side of the house.
Suddenly, however, the motion appeared to change, the crack in the south wall
ceased to open and on passing round to the western end of the house, the
building appeared now to be swinging from north to south and vertical cracks
began to open over windows and along lines of weakness in the west wall, but they
did not gape to nearly the same extent as the large crack in the south wall.
Nevertheless, judging by the fallen portions of other houses, the greatest damage
was due to this second series of vibrations, for at No. 3, Theatre Road, parts
of the parapet fell from the north wall. At No. 9, Harrington Street (Plate XXIII,
fig. 1) the parapet onthe south wall (over the porch) fell, while the fall of the north
wall of No. 8, Circular Road, has already been mentioned. Other examples of this
are No. 6, Badeapara Lane, Bhowanipur, where an old two-storeyed house in the
bazar has partly fallen, chiefly to the north! and the building occupied by the
Asiatic Society, a skylight on the roof having been tilted through an angle of
5 tothe north. At No. 6, Alipur Lane, belonging to the Raja of Burdwan, the
porch, which is on the north side of the house, has come away from the body

1About 80 yards further to the south, howe ver, the house at the corner of Badeapara Lane
and Shambhu Nath Pandit Road has lost the south «west corner which fell out to the south-
west.

E278. ?

APPENDIX A: H. H. HAYDEN. 279

vf the building, and the massive brick pillars supporting it have bulged out near
the base.

Examples of falls to east and west, however, are not wanting, the best being
‘seen in the Baptist Chapel, No. 44, Circular Road, already mentioned, and the
house No. 38, Chowringhee, from which much of the western parapet and a
verandah on the west side of the house fell.

There is little doubt that the fact that greater damage was done to north and
south walls is due to the peculiar construction of Calcutta houses; for the
majority are so oriented as to obtain as much of the south breeze as possible,
and consequently they are in most cases considerably longer from east to west
than from north to south, and are therefore more stable in the former direction.

Calcutta to Darjiling.—Eastern Bengal State Railway, and Darjiling and
Himalayan Railway.

No damage has been done at Bagula beyond cracks in the houses,

۱ but at Krishnagar many houses in the bazar were
rich buen badly injured, though few actually fell. In the
y injured, g y

European quarter, the Collector's house and the house of the District Super-
intendent of Police were bad!y damaged, while a considerable part of the Joint
Magistrate's house fell down. The east gable also fell out of the kacherz, but the
Circuit House, which stands only about 50 yards away, was scarcely even

cracked ; this was due to the fact that it was a low building and fairly new.

The church of the London Missionary Society suffered severely and the steeple
fell. This steeple, which was over the porch at the west end of the church, did not
fall en bloc ; portions appear to have been shaken off at various times, falling in
ten different directions, v2z.—

S. 273? W. 5.9. S 48° W.
S. 12? E. — N. gj: W. N. 133 E.
Ns 335 E. E. 224° N, E. 274° S
E. 38° S.

The two largest masses fel to E. 273? S. and E. 225° N. respectively.

It would thus seem that the steeple had vibrated in almost every direction ;
but it is more probable that these directions are merely modifications of two
principal vibrations at right angles to one another, the one being about NE to
SW and the other SE to NW :.and that partly owing to a resultant elliptical

motion and partly to unequal resistance at different points in the masonry, ap-
parent irregularities in the direction of overthrow were introduced.

At Chara, a village 11 miles north of Krishnagar, the church steeple also
fell : the main portion fell to S. 421? E., while smaller
portions fell in various directions. Pinnacles also from
the corners of the church fell E. 52° N. and E. 32? ۰

After leaving Bagula no striking effects of the earthquake are seen till
reaching the Baral river. Here the railway bridge has been much damaged ; the
massive hexagonal brick piers in the middle of the river
have sunk and lean over to the south-west. In one case the
sinking amounted to 113 inches and in another to 7 inches. The rails on the bridge
were displaced to the west; when seen by me, they had been to some extent

(279 3

Chapra.

Baral Bridge.

280 OLDHAM : GREAT EARTHQUAKE OF 1897.

straightened, but the engineer in charge informed me that the displacement had
amounted to 3 feet 10 inches, while the sinking of the piers had converted the
lineinto a switch-back?,

From this point onwards damage had been considerable and was noticeable

Ari? everywhere,and on reaching the Atraz river fissures were
tral. à
seen for the first time; here they run almost due east and
west, parallel to the banks of the river, and continue for several hundred yards,
with a breadth of (on an average) 6 to 8 inches,

At Atraí railway station the buildings were badly cracked, and portions of
the houses had fallen.

At Raninagar the ground was much fissured ; the fissures were often three

è hundred yards long and ran parallel to the railway, and the
Raninagar. : :
T rails were frequently displaced.

In this neighbourhood numerous hutscollapsed, chiefly towards the east
They were built of bamboo and thatch, and the bamboo posts had in most
instances broken near the centre.

At Sultanpur the railway station and other brick buildings have suffered,
gable walls and corners having fallen. The south
western corner of the railway station fell to the south.

At Tilakpur several bamboo huts collapsed, but the railway staticn, which

Tilakpur. was a low, squat building, ard apparently new, was
uninjured.

At Akhalpuv the north-west corner of the railway station fell to east-south-

-Akhalpur. east, carrying with it part of the roof of the verandah.
At Saidpur buildings suffered badly : several houses were badly cracked and
$ portions of walls fell, but no good indiĉations as to the
See: direction of the wave were obtained.

Three miles beyond Saidpur the fissures were very marked. They were seen
crossing the railway transversely, running north and south for many hundred
yards, and often having a width of over 2 feet.

About two miles beyond Darwanz the ground is much fissured, and the gate»
house at telegraph post No. 129/22 is completely wrecked, having fallen to the

Sultanpur.

east.
At Nilphamari uut station the fissures run north and south, and are often

e ] as much as 10 feet wide, while areas of from 40 to 5o .

A feet long by 1o feet in width have subsided to a depth

of 3 feet, and in some cases nearly 4 feet. The goods shed on the eastern side of

the line was badly injured, the southern half of the floor having subsided. In places
the ground beneath the rails has sunk, and rails and sleepers remain suspended.

Similarly at Zaldibari railway station the permanent way and platform are
cut up by fissures running north and south and a large

10۰ portion of platform has subsided. Between the railway
station and the dak bungalow there are fissures and more or less circular holes
trom which, it is said, sand and water were ejected during the earthquake.

1 The bridge runs north and south.

( 280 )

APPENDIX A: H, H. HAYDEN. 281

After leaving Haldibari the line is very badly damaged: at telegraph post
161/8 the gate-house is cracked and the south wall leans outward, gaping 7 inches
at the top and about 1 inch at the foot, being displaced 5° from the vertical. The
fissures -here run from S. 18° W. to N. 18° E. Beyond this, deep
fissures, varying from 3 to 7 feet in depth, run along each side of the rails: asa
rule they are from 6 inches to 1 foot in width, but in places wider, in which case
the rails and sleepers are supported only by a narrow column of earth running
along the centre of the permanent way.

The rails also are often displaced, the displacement being usually towards the
west. In some cases the displacing force has been sufficient to break the solid
steel rails, but, more frequently, fracture has taken place at the junction of the rails,
the fish-plates being broken and bolts torn out.

At about half way between Haldibari and Mandalghat four telegraph posts
have fallen, and point in the direction W 10° S: this observation is, however,
of no great value, as the direction of fall must have been largely modified
by the pull of the wires. Here fissures run transversely to the railway (S. 10°
E, N 10° W), but gradually draw away from the neighbourhood of the line
and strike across the open country in a direction S "و‎ W, N 9? E.

At about 3 miles to the south of Mandalghat the fissures gradually begin to
disappear, and at Mandalghat railway station they are
no longer seen. Between this station and Jalpaiguri
small overhanging banks of drains and ditches have collapsed, but no true fissures
are seen.

At Falpaigurt some little damage was done: the post office, a two storied
brick building, was rather badly cracked, while some of
the inner walls of the new club, which is in process of
construction, were damaged, large cracks having opened in directions at right
angles to one another and at 47” to the horizon.

The only actual falls, however, took place at the District Board Office. This isa
rectangular building, 24 feet high, and with a flat roof, round which runs a parapet
4' 5'in height; at each corner the parapet was surmounted by a solid cupola ;
two of these cupolas fell.

One fell in a direction S 18° E and stands upright on the ground exactly as it
originally stood on the parapet. .

Original height of base above ground =24 feet.

Horizontal distance through which it was projected=Sft. 6in. from base of wall.

The other cupola fell N 23° W and lies on the roof at an horizontal distance
of 9 feet from its original position ; the height through which it fell — 7.e., height of
parapet from roof=4ft. sin. It is probable, however, that it rolled over after it
fell.

At Belakoba railway station the eastern wall of the station house is cracked
at its north-east corner, while other small brick buildings
in the neighbourhood have been cracked, but no serious
damage appears to have been done.

Between Belakoba and Siliguri fissures again appear at telegraph-post 192/12 j
here they run parallel to a small river and cross the railway at right angles

( 281 )

Mandalghat.

Jalpaiguri.

Belakoba.

282 OLDHAM: GREAT EARTHQUAKE OF 1897.

They are apparently only 3 to 4 inches wide, and are purely local, disappearing
again immediately, nor are fissures again seen till some miles beyond Siliguri.

This absence of fissures between Mandalghat and Siliguri is no doubt due in
great part to local peculiarities in the soil, which, from being of a light sandy
nature in the fissured region, gradually assumes a dark loamy character, and it
would appear highly probable that the greater tenacity of this latter enabled it ta
withstand strains under which the porous and ill-compacted sands gave way. A
similar absence of fissures for a considerable distance along the railway to the
south of Kuch Bihar will be noticed further on (p. 286).

At Siliguri no damage appears to have been done, in fact all signs of a
violent earthquake are strikingly absent: the railway
station is slightly cracked, but dces not appear to have
suffered any damage beyond the loss of a few loose bricks; and it is not till the
Mahanadi river has been crossed that we come on any marked effects of the
shock, Between that river and Sukna Railway Station, however, the ground is
fissured in several places, the cart road which runs beside the railway line being
much cut up.

Darjiling and Himalayan Railway.—Very little damage was done at any part
of this line, though many of the bungalows and tea factorieson the neighbouring
spurs suffered considerably. As already stated, fissures opened between Siliguri
and Sukna.

At Tindhavia railway station the station building was cracked and some

stones fell from the gables, while two small landslips
Tindharia. occurred on southern hill slopes. A larger landslip
occurred subsequently ( June 24th ) at this station, and

much of the hillside below the station master’s house was carried away.

Darjiling.—Here the most striking effects of the earthquake are seen in the
large numbers of chimneys which have fallen. Many houses, which were other.
wise uninjured, have been badly damaged by the falling
in of the great heavy chimney stacks which crashed
through the rocf, breaking everything before them. As a rule, however, they are
not of much service in determining the direction in which the wave travelled.

A chimney on Collington fell to the N 8° E while the western chimney
on Dr. Anderson's house on Jalapahar fell E 8° S.

At Snowy View, the front of the house fell out to the north, while most of the
inaner walls also fell, and a small window inthe east wall fell inwards in the
direction W 28? N. A small cottage afew yards below Snowy View was
completely wrecked : the walls collapsed and the roof fell to the W 28° N
while a window in the east wall fell out E 38° S.

Singamari was a new house built of stone and mortar. The upper part of the
western gable wall fell out and several inner walls suffered badly. Cracks ran
through the house from N 32° W to S 32° E.

Inthe old Bhutia Cemetery, on the easterm side of the Ridge and just below
the Calcutta Road, several old tombs were partially thrown down.. These
were built chiefly of loose stones, uncemented by mortar, and having on their
summits small pyramidal heaps of smaller stones, or single ovoid or conical stones

( 232 )

Siliguri.

Darjiling.

APPENDIX A: HH. HAYDEN, i
propped up on end. Here one tomb fell over to the E 38? S, while the top-
stones of several others are tilted to E 38? S and E 28? S,

In the European Cemetery
on the western side of the hill,
below the Shrubbery an obelisk
has been affected in a very
interesting manner. At the
plane of junction between two
of the stones, the plaster has
given way, and the upper por-
tion of the obelisk has been
disturbed in such a manner
that it has apparently been
twisted through an angle of
about 2°: as shown in plan in
figure 31. The monument is
built of sandstone, of which a
small chip gave a specific
gravity of 2:485; this givesa
weight of 123 cwt. for the upper

Fig. 31. Plan of twisting of obelisk in the Cemetery at part of the monument.
Darjiling.

As will be seen from plate XXXII, fig. r, the chief displacement occurred at
the third joint from the apex,a slight displacement also occurred at the next
higher joint, but was too small to be measurable.

[In any calculations based on the displacement of this monument the factor o
the adhesion of the plaster may be neglected, as the layer between the stones
was not more than ak inch in thickness, and has now crumbled almost into dust. ]

Besides the above monument a few tombstones have been affected ; the ground
appears to have been shaken loose, and the tombstones have sunk on one side
and are now thrown out ofthe vertical, but this is not very marked.

The most useful observations asto the direction of wave were obtained from
the old chimneys at Senchal, near Darjiling. These chimneys are all that now
remain of the old barracks; the intervening walls
have completely disappeared,and the chimneys remain as
isolated towers. One of these fell completely to E 28° S : the mass of the chimney
(which was originally 20 feet high) now lies along the ground up to a horizontal
distance of 35 feet from the base of the chimney, but smaller stones and mortar are
found scattered about insome quantity up to 51 feet, while one piece weighing
olbs. was found ata distance of 78 feet. The piece appeared to have fallen where
it lay, for the ground under it was dented and the grass killed, while the moss»
with which the chimney was thickly covered was still found adhering to the stone
which would make it seem highly improbable that it could have rolled.

A few small stones fell to W 28° N at thé foot of the chimney. None of the
remaining chimneys collapsed completely, but portions fell due several of Ee
and all agree in indicating the same direction of wave, viz., E 28° S, to W. 28? N.

ee. Vi)

e

DAY W, laa"

Senchal.

284 OLDHAM: GREAT EARTHQUAKE OF 1897.

Kuch Bihar Railway.—Turning now to the railway from Parbatipur to Kuch
Bihar, the effects of the earthquake are everywhere
striking. At Shampur, the ground is much fissured
and the station platform has sunk, while alorg the side of the railway the stone posts
used in wire fencing have in many cases been snapped off at various heights above
the ground : some almost at the ground, but the majority have been broken off
within a foot or so of their summit.

These posts are made of sandstone and are usually about 3 feet high, 4 to 6
inches wide and 2 inches thick : the wire passes through holes in the stone. Asa
rule, the posts have not broken at the holes. The fracture is probably due to jerks
and pulls of the wires as the waves passed, for it would seem impossible for such
short and substantial posts to have vibrated sufficiently to produce fracture.’ .

Near Shampur also, the hexagonal brick piers of one of the bridges have been
broken through horizontally, and the upper portion has shifted slightly ; this
form of fracture appears to have been rather common, and though at first supposed
to be of no great moment, was subsequently found to render the bridges unsafe
for either rapid or heavy traffic.

Ofall the places visited by me, Rangpur was undoubtedly that which had
suffered the most damage. The railway station was not
very seriously injured, but in the town itself, which is
situated some 3 miles west of the railway station, almost every brick building was
irretrievably damaged, while several were almost'completely overthrown. A few,
however, escaped with only the loss of the qus story, the lower portion being
badly cracked, but possible to repair.

The damage was not due so much to the actual shock as to the secondary
effects of the earthquake : for the whole town being built on a loose and porous
sand, overlying, at a depth of only a few feet, a substratum of waterlogged sand,
the surface is everywhere cut up by fissures, which run parallel to main lines of
weakness such as canals, sides of tanks, etc. These fissures have frequently
opened under houses, in which case there has been differential subsidence and
consequent rupture of the walls. The directions of the fissures are chiefly from
E 4°—10° Sto W 4?—10? N and N 4*—10? E to S 4°—10° W though some run in
other directions. The first direction, however, is the more noticeable owing to the
fact that it is roughly parallel to the canal, which runs through the town, and
is, of course, a line of weakness.

Near all tanks the fissures run parallel to the banks, and consequently intersect
at right angles, and it is owing to their proximity to alarge tank that the
most important buildings have suffered so severely : in fact, some of them stand
between a large tank and the canal, and are completely undermined bya network
of fissures, resulting ina differential movement of adjacent parts of the buildings
and consequent destruction. This effect is well seen in the Judge’s house and
inthe courts. (Pl, XXV, fig. 1.) In the case of these two buildings the
chief destruction has been caused by the same fissure, which has also undermined

Shampur.

Rangpur.

1 On comparison with the effect of the earthquake in the epicentral tract of the Khasi and
Garo Hills it would seem that the explanation, here rejected, might be accepted. —R,D.O.

( 284 )

APPENDIX A: H. H. HAYDEN, 285

numerous other houses in its line. Houses, however, which have not been thus
built near lines of weakness have in several cases escaped with a comparatively
small amount of injury, while the Raja of Dimla’s house, about 4 miles to the east
of Rangpur, was almost completely surrounded by tanks and consequently suffered
very badly, so much so, indeed, that the Raja appeared to have given up all idea
of attempting to repair or rebuild it. Indications of the direction from which
the shock came were obtained from two plaster sowars* which fell from the gate-
posts at the entrance to this Raja's palace : both fell E 64? S.

In the cemetery at Rangpur two similar monuments were partially overthrown,
but unfortunately they were close together and both fell inwards, the space between
the two being filled with a confused mass of bricks and mortar, it being impossible
to decide from which monument any particular portion fell; consequently no ob-
servations as to velocity could be obtained ; the directions in which the monuments
were overthrown were, however, E. 7? S. and W. 7° N. Other monuments
were much tilted owing to sinking of the ground in some parts more than in others :
this effect was seen to a greater or less degree in every cemetery visited by me.

The following is a list of the observations which give indication as to the
direction of the shock :—

Monuments ir cemetery ; Mis MM „E. 7 S. and W. ? N.
Pillars on east side of library verandah e » E. 419 S.
Gate-post at entrance to Collector's house e e, 47 S.

This pillar fell mainly into two blocks and many loose bricks. The upper
portion weighing about 7 cwt. (calculated) was completely overturned and dropped
down almost vertically at the base of the wall; it lies W 43” S of the base
of the gate-post. The next piece, though it actually lies to E 43^ S of
its original position, nevertheless fell, together with the top, towards W 43° N.
In fact, the pillar was broken off at 2ft. 1oin. from the summit : the lower part of
this fell to E 43° S, while the upper part was split into two, of which the
topmost was overturned completely and the next portion only partially so. That
is to say, the pillar broke while vibrating towards E 43? S, but owing to
its inertia the upper portion was left behind and at the same time overturned, and
points towards W 42° N.

The canal being, as already stated, aline of weakness, it is not surprising to
find that the banks on each side are cut up by fissures, while its bed has risen, in
some cases through several feet, the central portion being now above the water :
this is well shown by the bamboo bridges which have been shot up in the centre.
(Plate XX VI, fig. 1.) The same effect is seen in numerous places between Rangpur
and Kuch Bihar, where bridges of small span cross canals, small water channels or
swamps. If the bridge has a central pier, then this pier has been thrust up and,
the bridge broken. This is, however, in some cases due partly to a sinking of the
abutments as well.

Similar to this effect is the silting up of wells which took place throughout
Rangpur district, These wells were of no great depth, water being obtainable at

1 Horsemen,

OE

286 OLDHAM: GREAT EARTHQUAKE OF 18979»

a few feet below the surface. After the earthquake they were found to be filled
with sand. The sand is the same as that which was thrown out of the fissures, and
is a rather coarse variety containing much muscovite. At Haripur,in Rangpur
district, rolled pieces of lignite were said to have been ejected in large quantities
with the sand: it is therefore highly probable that the sand was originally derived
from the disintegration of the coarse tertiary sandstones which are seen in large
quantity on the lower slopes of the nearest range of the Himalayas.

Sand was thus ejected in every part of Northern Bengal visited by me;
and the quantity thrown up must have been enormous; for lines of sand, four or
five yards in width, are often seen running almost uninterruptedly for miles
parallel to the railway. In this sand occur numerous vents, either in the form of
conical hollows or of small cones built up of sand to a height of several inches.
These vary in width from a few inches to about 10 feet : the latter size is rare, the
more usual width being from 1 to 3 feet. These are well seen on the Kuch Bihar
branch between Teesta and Dewan Hat. The amount of sand ejected is variable
but in some places it covers the ground to a depth of over a foot, as on the chur
between Moghal Hat and Gitaldaha, where the railway has been buried for about
200 yards under some 8 inches of sand.

Between Rangpur and Kaunia the railway line was much damaged; the
permanent way was fissured and, in places, completely undermined. Bridges are
broken, that over the Monas river having been rendered unfit for traffic. This
bridge consists of three steel girders supported on
hollow cylindrical iron piers, and resting at each bank on
brick abutments. The fissures which opened along the river bank completely
destroyed the abutments, while the piers have been tilted as much as 15° from
the vertical, and the ends of the girders no longer rest on them.

All along this line, the distortion of the rails is very marked : they have been
displaced and bent in an east and west direction : this distortion has usually been to
the west, but ina good many cases it has taken place in the opposite direction:
In a north and south direction, the ground has been thrown into a series of folds and
undulations ; the whole effect being that of an enormous ** switch-back." Places
formerly level have,risen or fallen, and differences of level, often amounting to
several feet, are found in the space of a few hundred yards. The effect of the
motion producing these undulations has been to develop pulls and thrusts in the
rails. Where they have been in tension, adjacent rails have drawn away from one
another at the ends, breaking bolts and fish-plates, and where they have under-
gone compression, the rails have been distorted, usually in the centre, and often
with such force that the rail has broken.

Returning now to the placesof interest between Rangpur and Kuch Bihar,
we find that at Kawnia much damage has been done:
the transhipment-inspector's bungalow was completely
wrecked, while the land in front of it was much fissured and sand and Waker rushed
out. An interesting fact as regards the difference of period of consecutive waves
is well exemplified at Kaunia: if wetake aline running north and south through the
bungalow just mentioned, we find that much damage has been done along that

( 286 )

Monas Bridge.

Kaunia.

APPENDIX A: H. H. HAYDEN. 287

line, but on a line at about too yards to the west a somewhat similar bungalow
has remained standing, and the other buildings along this second line have come
off without much damage: further west again, we find a line of greater destruction
and it is probable that, if this were followed up, we should find analternating series
of lines of destruction and of comparative immunity. Even allowing for the varie
able factor necessarily introduced by the difference in construction and age of the
various buildings, the series of lines above mentioned is sufficiently marked to
make it a matter of great probability that the destruction or escape was to a large
extent due to interference.

Passing on from Kaunia towards Kuch Bihar, much damage has been done
all along the line up to Chaarha Hat. For some 5 or 6
miles beyond this station, however, the effects of the
earthquake are hardly perceptible, the absence of fissures being as marked in
this area at it was between Mandalghat and Siliguri. The soil, too, bears
some resemblance to that of the latter area, but is rather more sandy.
Itis, however, not improbable that the absence of fissures may in some part
be due also te the large numbers of bamboo clumps throughout the surrounding
country: as far as one can see on either side of the railway, the country is
covered with these clumps, which, by their thickly matted roots, bind the earth
together and thereby prevent the formation of fissures: but as the railway itself
is barely damaged, it is more probable that the soil is chiefly accountable for this
immunity.

Chaarha Hat,

In the neighbourhood of Dewan Hat, however, the line has suffered severely
and the bridges, particularly that over the Manshai
river, have been broken. At about 7 miles south
of Kuch Bihar a small bridge passing over a water channel in swampy land has
been damaged by the thrusting up of the central pier. At this spot also the sur-
face of the surrounding country has been covered with sand and water, from
which in many places bubbles of gas are seen rising: this gas has an unpleasant
odour somewhat resembling both H,S and SO,, but though it was asserted by
many people that the opening of the fissures during theearthquake was accom-
panied by a * smell of sulphur * which was popularly accredited to volcanic pheno-
mena, it is probable that these gases are merely derived from the decaying vegeta-
tion which has in so many places been covered by the sand and water. This
is the more probable as the same ebullition of gas may be seen taking place in
small puddles in the fields in districts such as Krishnagar where no fissures
were formed.

At Torsa Railway Station the station-house has not suffered, but the banks
of the river have been much cut up by fissures on both sides, and have in many
places caved in.

At Kuch Bihar most brick buildings have been severely damaged. With the
exception of the Palace and the Superintendent’s house, the principal build.
ings surround a large tank, and have conse-
quently been injured by the fissures which have
opened in their neighbourhood, 9 enough, however, the building which

( 387. -)

Dewan Hat.

Kuch Bihar,

28 OLDHAM: GREAT EARTHQUAKE OF 1897.

in the whole town has suffered least is built on the western bank of the same
tank. This building is the Dewan’s house, but it owes its escape to the fact
that it is builton a raised mound of earth which apparently acted as a cushion
and enabled the house to escape with only a few cracks and with the loss of one
small cupola from the parapet.

The Palace was rather badly injured, but chiefly by the fall of numerous
immense turrets which carried with them parts of the adjacent walls, and often
crashed through the roof carrying all before them.

The Superintendent’s house was seriously injured, in fact ruined, by a fissure
Opening beneath it, with the result that the building was broken intotwo. It wass
however, like many of the other buildings, built on sand, which in its turn lies
on another bed of soft water-bearing sand (almost a quicksand).

The chief directions of overthrow at Kuch Bihar are :—

: : : S42 ES
W 28? N.
EOS US.
E 231? S.
E 20° S.
A marble vase in front of the palace fell . . E 209 S;
Lansdowne Hall.—Clock tower fell . : ESO
Superintendent's house.—Urn from turret fell . E 20° S.

Palace turrets fell . A

Brick gate post . 5 کک وہ‎ SRE.
A tall brick watch-tower in the town fell . JUN. 9» E.
Minaret of a mosque . ; . : . W 28? N.

Between Kuch Bihar and Alipur Duars the road is fissured and bridges
broken, while at Alipur the western wall of the gaol fell. Between Alipur and
the foot of the hills not much damage has been done, as

the country is mostly protected by forest, but at Baxa‏ کر ہہ
(about 2,000 feet) all stone buildings were damaged, while many collapsed com-‏
pletely. The hill sides are everywhere scarred by landslips, and much of the‏
road between Santrabari and Baxa was carried away. Owing to reflection of‏
the waves, objects fell in every direction, but the prevailing direction at Baxa and‏
the neighbouring picquets was E 48° S. Near the foot of the hills, however, a tree‏

fell to W 38? N.

3. Report by the late Mr. G. E. Grimes, Assistant Superintendent, Geological
Survey of [ndia.

CALCUTTA, EASTERN BENGAL, CACHAR.

Calcutta.—At the time of the commencement of the earthquake I was lying
down reading on a sofa in aroom on the top storey of No. 11, Middleton Row
and the first indication of the shock which , I perceived was a loud rumbling noise
and a staking of the wooden partition, as if something very heavy, much heavier
2 tons 11 cwt. 20 lbs. It fell from a height

! The weight of this turret was calculated to be 1
iter ce of 26 feet from the base of the wall,

of 38 feet, and was found lying at a horizontal distan

Cat

APPENDIX A : G. E. GRIMES. 289

than the floor would have supported, was being rolled along the corridor just outside
the room. I started up, but could not at first make out what was the matter, and
it was only when I got on my feet that I felt the ground rocking underneath me.
After rushing out of the house I stood in the compound on the west side, and
viewed from that direction, the house appeared to be rocking in a north and south
direction. A few minutes after the earthquake was over, I went into the house to
see what damage had been done, and I then saw that the north-east and south-west
corners of the house had suffered more severely than the other parts, and that
there the walls were leaning out more than in the other corners. Besides this,
the verandah on the south side of the house, which was built at a later time than
the main building, was separated from the rest by an east and west crack.

On the following days I went round parts of Calcutta, chiefly in the area
between Park Street and Bow Bazar Street, and also to Howrah, to examine the
nature of the damage done. Within this part of Calcutta there are a large number
of native bastis, the cutcha huts of which show practically no signs of being shaken
by the earthquake, but scattered amongst them there are a large number of pucca
houses, many of a considerable height, and most of them showing signs of age and
dilapidation, but despite this the damage done was not so severe and prominent
as immediately to the south of Park Street, where the houses are almost all detached
and lofty; many of these are old and show signs of very frequent patching up.
I made the following notes on the chief buildings affected by the earthquake : —

At St. Thomas’ Church, Free School Street, there are two large cracks and
several smaller ones in a north and south direction ; by one of these larger cracks
the chancel was separated from the main body of the church, and by the other
Crack the porch was separated from the church. In the Boys' Free School, which
is close by the church, the cracks were at right angles to those in the church, the
facade with Doric pillars of the School, which faces to the south, being damaged
by east and west cracks. From the top of the spire of the Free Church of Scotland,
Wellesley Street, a considerable piece had fallen, and the next day I saw it on the
ground, considerably broken, near the base of the tower, and taking the position of
the largest and main block of the fallen top of the spire as the position where the
top had struck the ground, and there was no evidence of its having rolled to that
position. I found that it must have fallen in a direction W 10° S. The tower
on which the steeple is built was separated from the main building by an east and
west crack. St. James’ Church, Lower Circular Road, had three small minarets fallen
from its steeple, but as the fragments had been cleared away, there was no evidence
of the direction in which they were thrown. The other damage to the church con-
sisted of several east and west cracks running along the length of the building.
From a house just off Wellesley Square, near St. Saviour's Church, the south parapet
wall had fallen into the yard below towards the south. A large pucca house at the
Circular Road end of Dhurumtollah Street was very badly damaged, and the front
(south) quarter of the house was separated from the rest by'a large east and west
crack, but the south-west corner of the house was much more damaged than the
south-east corner, No. 114, Lower Circular Road (Plate XXIV, fig. 1), was very
badly damaged, and the south bay of the top story had fallen out towards the south.

U ( 289 )

290 OLDHAM: GREAT EARTHQUAKE OF 1897.

At Messrs. Traill & Co.’s business premises, at the back of the Great Eastern Hotel,
the front portico and balcony which face to the west were completely destroyed, and
at the back the eastern parapet wall had fallen over towards the east, and, after
carrying away the bathroom below, had crashed through the roof of the godowns.
The front part of the house in Bentinck Street, opposite the entrance to Mangoe
Lane, had fallen forwards into the road (Plate XXIV, fig. 2); this side of the house
faced the east, the front wall being builtin a direction N xo? E—S 10° W.

After crossing the river to Howrah we see that here, as in Calcutta, a few houses
have suffered very severely, whereas others quite close to them have eseaped with
little or no damage. The north-west corner of one of the houses facing the Howrah
Green has fallen out, and the direction of the slope of the rubbish is to N 25° W*
The eastern parapet wall of the next house has fallen down. The only
other badly damaged house I was able to examine was that of the Deputy
Magistrate of Howrah in the Grand Trunk Road. This was very badly cracked
everywhere, but the south-east corner had suffered most and fallen outwards.

„ At the time of the earthquake the Chaplain of Howrah was watching the water
in the tank near his house, which was swaying violently to and fro, and the
direction of this movement, which he pointed out to me, was almost exactly north
and south.

Looking over the notes above, we see that there is very little definite evidence
with regard to the direction of the earthquake waves, and nothing from which one
can calculate its intensity. In Calcutta the fallen walls have mostly gone over
towards the south or west, and the corners of the houses in Calcutta which have
suffered the most are the north-east and south-west corners, and this would seem
to point toa NE and SW direction for the earthquake wave, with variations
from a direction N and S to one E and W. In Howrah, however, the evidence
rather points to a NW and SE direction.

Dacca.—At Dacca the earthquake was felt much more severely than in
Calcutta, and the relative damage is much greater ; almost all the pucca buildings
in the place being badly damaged, some have entirely collapsed and very many
were uninhabitable. | ;

For the direction of theearthquake wave the following are the chief pieces of
evidence. In almost all the buildings the greatest damage is that done to the
north-west and south-east corners ; and in some houses as the club, dak bungalow,
etc., the south-east corner has fallen out. The top half of one of the low pillars
round the race-course has been broken off and thrown from the base, and is now
lying on the ground at the side, the direction in which it has moved being S 25* E. .
The top ornament of another pillar is broken off, but part of itis left hanging
on its centre rod, which is bent, and the direction in which it has fallen
is W 15° S. Alsoonone of the gate pillars the top is bent in a direction
N ro? E. In the centre of the race-course there is a Hindu temple with
a lofty narrow spire ; the top of this is broken off, shifted toone side, and bent
over, but it was prevented from falling, Lam told, by a strong iron rod through
its centre ; the direction of movement of the top I determined to be as near as
possible S 20° E. The Nazir’s mutt, a Muhammadan shrine on the banks of
the river, has been completely overturned in a direction which, as near as

( 290 )

APPENDIX A: G. E. GRIMES. 291

I could make out, was N 25° W. In the Armenian churchyard there is
a lofty square tower, the walls of which are set almost exactly north and
south, east and west (N 6° E, and at right angles): the whole of the
northern half of this tower has fallen over towards the north. From the
European cemetery but little evidence is obtainable, as only two or three tombs
showed any signs of damage. From the tomb of John Charles, infant son of
T. Richardson, Judge of the Civil Court of Dacca, the top is broken off, but most
of it is still held onto the central rod, which is bent over to one side, and the direc-
tion in which it is bent, and the fragments have fallen, is E 10” S. The
only other tomb which shows any indication of the direction of the earthquake
is the very large and lofty one which is ascribed to * Columbus saheb, kampani ka
naukar” ; this had on its sides, at three different heights, small minarets, some of
which have fallen ; little, however, could be ascertained from these, asthere was
evidence to show that in almost every case the top ones in falling had struck those
lower down : besides this, the fallen pieces had, previous to my visit to Dacca, been
shifted from their original position, and although the marks where they had struck
the ground were visible, it was impossible to tell which fallen minaret had made
them ; one, however, of these minarets had apparently fallen free in a direction
W 10° S. The Engineer Babu at the water-works told me that at the
time of the earthquake he was standing close by the tank there, and that the
water in it was swaying to and fro ina direction which I measured, when pointed
out to me, as S 15?—20? E.
Putting the above ina tabular form, we see that—

Club and Dák Bungalow . indicate NW and SE.

Low Pillar, near Racecourse . a - Nas VW, ace 394 E.
Second وو ” وو‎ ^ ew E15 N; » ۷ 8ہ‎ S.
Gate ,, ah 3» 3 9 Nue B; y» S10 W.
Hindu Temple on Racecourse 5.» Na Wa ei 5.20” E,

Nazir's mutt S . g ری‎ NIIS NW npo کا‎ E.

John Charles Richardson's Tomb . , W 10? N, E RO پک‎
Columbus Saheb’s Tomb ; «abo RE ht e ewe NO E
Tank at Water-works . e » W15°—20°N,, S 15 —z0" E.

Looking at this evidence, and also considering that the impression of the
inhabitants of Dacca was that the direction of the earthquake-wave was along
a line north-west and south-east, we see that the most reliable observations point
toa direction N 20°—25° W.

Of the velocity of motion of the wave-particle the pillar near the racecourse
is the only thing which gives any evidence.

Before the earthquake this pillar, of which the horizontal section is square, was
2 feet و‎ inches high, and the length of one of its sides 1 foot 83 inches. From this
the top part, 1 foot high, had been thrown off ina direction S 25° E, and so
the height from which its centre of gravity had fallen was 1 foot g inches; this
piece was standing upright and perfectly square with the bottom part, the
distance of its centre, from the centre of the basal piece, being 3 feet 2 inches.
At first I felt doubtful if this had been thrown off by the earthquake, but an

00 ( 381 ^)

^

292 OLDHAM: GREAT EARTHQUAKE OF 1897.

examination of the grass underneath showed that it had not been long in its
present position. The pillar was old and considerably weathered, and along the
surface of the course, where the parting occurred, the weathering had almost
entirely loosened the top part from the bottom, so that there was very little cohesion
between them. |

For the time and duration of the shock the only evidence at all reliable was
that obtained at Dacca by the Traffic Superintendent of the Dacca-Mymensing
Railway, who looked at his watch directly he felt the earthquake. He tells me
that the shock started at 5-11 P.M. local time, and lasted almost exactly go seconds.

Dacca-Mymensing Railway.—Of the damage done to the railway line
between Dacca and Maimansingh I was not able to see much, as the line had been
largely repaired and the trains were running again when I travelled over it. 1
could, however, see that the most serious damage to the line had been at its
northern end ; for the first two-thirds at least of the journey I saw practically no
damage, and a great part of this was through dense jungle growing on the reddish
lateritic beds of the older alluvium at the east side of the Madhupur jungle.
During the latter part of the journey, however, the railway was running on the
newer alluvium, and it was here that damage was done and the lines curved
and bent by the earthquake ; this curving and bending had been entirely removed
before I got there, and the only evidence of it was some bent rails lying at the side,
and these showed that the shock must have been very severe. Across the line and
running through the fields on each side, there were very many fissures in the
ground trom which sand and water had been forced up and spread over the
ground on both sides of the fissure. This sand, when I saw it, was of a light
bluish-white colour, but 1 was told that, when it first came to the surface, it was of a
blackish colour, and had a strong sulphurous smell. Tothese fissures and the
sand from them I shall refer later on. This portion of the line was saved from
more serious damage by its running along the surface of the ground ; if it had
been on raised embankments, like a great part of the Assam-Bengal Railway;
these would have been most probably fissured and overturned, the bridges over
the streams been broken and buckled, and the damageto the line much greater
thanit was.

Maimansingh.—At this place the earthquake was very severe, and the damage
done was enormous. Almost all the pucca buildings were atleast partially wrecked
and many were complete ruins. Of the church, only a small part of the wall at
one end was left standing ; one of the Hindu temples has fallen completely down ;
the upper part of the Raja h's, Palace had all fallen, and the lower story was very
severely damaged ; the upper part of the Judge's cutcherry had likewise gone,
the Judge's house also; one end of the Collector's eutcherry was in ruins, and,
besides these, many other places too numerous to mention, including the bazar,
which was a complete wreck.

In Maimansingh the collapse of a structure is usually so complete and the
different parts interfere with one another so much that indications of the direction
of the earthquake-wave are rare, and then are mostly somewhat vitiated. The
best indication of the direction is tke following. The top of one of the Hindu
temples (Deota Din Siwala) in the bazar has fallenina direction N 48° E

( 292 )

APPENDIX A: G., E. GRIMES, 293

and was lying at the side ; it had not, however, fallen free, for there was a mark on
the side of the tower where it had struck it, but one could see that the direction
of the horizontal component of the movement of the fallen top was not much
altered. Besides this, none of the other buildings show any accurate evidence
of the earthquake ; one of the gate pillars of the Rajah's palace compound
was twisted and shifted slightly in a direction N 15? E, but as there is
a heavy iron gate attached to the pillar, the latter could not have moved freely.
Corroborating evidenceof the direction deduced from the Deota Din Siwala
is, howeyer, obtainable from the cracks in the open country; these, of
course, vary very greatly in direction, but the ones which 1 chose are those
which were most out in the open country, and which would have been least
interfered with by irregularities in the surface ; these were running in a NW to
SE direction, and so the direction at right angles to them, which would
in this case be presumably the direction of the earthquake-w ave, was
NEto SW. At the time of the earthquake the north-east corner and part
of the outer walls of the jail fell down, the measurements of which from the corner
were 105 feet 8 inches of the wall built in a direction E 30° S and
46 feet of the wall at right angles ; if we join the ends of the broken part by a
straight line and take a direction at right angles to this line, we find it works out
N 54? E, which is not far from the N 48° E, which I obtained from the Deota
Din Siwala ; this may be only a coincidence, as in such a case as this it is not safe
to assume that the line of limit of the breaking of the wall was parallel to the
crest of a wave. Beyond the above case I was unable to obtain any evidence
of the direction of the earthquake wave, but 1 think I may safely assume that it
was, approximately, N E. to S W.

For obtaining the velocity of the wave particle, the only thing we have to rely
on is the overturning of the jail walls, which are ro feet 8 inches high, and 8
inches thick, the overturned portion being 105 feet 8 inches of the wall in a
direction E 30° S, W 30? N, and 46 feet of wall in a direction N 30° E, S 30° W.

For the time and duration of the shock the best information I have was
obtained through the Traffic Superintendent of the Dacca-Mymensing Railway.
He told me that one of his guards, a Sibpur boy, who was at Maimansingh at the
time, took out his watch directly he felt the shock, and it was 5-10 P.M., local
time, and the shock lasted almost exactly 9o seconds.

In the station of Maimansingh, and in the country around, the ground is fissured
in very numerous places, the cracks near the river being mostly parallel to it, but
not in all cases; in the surrounding country, however, they are in all directions,
following the lines of weakness. From most of these cracks sand and water, like
that seen near the northern part of the Dacca-Mymensing Railway, has been
forced out and has flowed over the fields, damaging the crops to a certain extent.

The fissuring of the country is accompanied, too, by sinking of the ground
often over large areas. More prominent cases of this may be seen in two or three
places on the road from Maimansingh to Muktagachha, where the cracks are nearly
at right angles to the direction ofthe road. In these cases the road is faulted, the
ground having sunk so that the road on one side of the crack is about 4 feet
higher than it ison the other. Wherever the direction of the earthquake-wave is

( 293 )

204 OLDHAM: GREAT EARTHQUAKE OF 1897.

normal or subnormal to the length of the road, and where the road is raised on an
embankment above the general level of the country, the cracks run parallel to
the road, and in this casethe action of the second semiphase of the wave on the side
from which the earthquake-wave is travelling, and the first semiphase of the wave
on the farther side, tend to throw the cracked sides of the embankment outward
and if these outward movements proceed tothe centre of the embankment, the
whole of it is lowered and there is a sinking in the road at that place.

Muktagachha.—T he town of Muktagachha was, I am told, largely built on made
ground, and here the fissuring and sinking of the ground was very severe, and
in consequence the Rajbari (Rajah’s Palace) and the pucca houses near it were com-
pletely wrecked. I have taken two photographs (Plate XXVIII, figs. 1 and 2)
from amongst these houses, The first is Moish Babu’s house, under which are
two large cracks along which the ground has sunk, and so the plinth of the house
is faulted by the two cracks in two steps, and one portion of the plinth has sunk
4 feet 6 inches below the higher portion; the crack under this house has a
direction W 15? S. The second photograph shows a more complete ruin, where
only a part of one of the walls of the house is left standing and this is leaning over
to the south owing to a large crack in the ground just on that side.

Maimansingh to Silchar.—From Maimansingh 1 went to Silchar, and on the way
from Narainganj to Silchar by steamer, I was not able to observe much, owing
to the rate at which we were travelling, a great part of the journey being by night
or in the early dawn. The state of the river, too, was not very favourable for my
observations, as it was in flood, so that the banks were covered up and it was often
difficult to tell if the flooding of the surrounding country was due entirely tothe
rise of the water in the river or partly to sinking of the ground. At many places,
however, on both sides cf the Barak, especially in its upper reaches, the
sinking of the land was quite evident, and roofs of houses just out of the water
were to be seen at several places, a prominent instance of it being at Karimganj
where a great part of the bazar was under water.

Silchar.—In Silchar station itself very little damage has been done, and from
the buildings there one can hardly tell that there has been an earthquake at all.
The bank of the river has cracked and slipped in places, but not badly, except
perhaps at the bend just above the station.

Practically the only evidence of the direction of the earthquake was obtained
from a cylinder seismometer like that described in Memoirs of the Geological `
Survey of India, Vol. XIX. This was kept locked up in a separate galvanised iron
shed close to the meteorological observatory, and was in proper working order at
the time of the earthquake. The cylinders were made of metal outside, filled with
cement ; they were 12 inches high, were numbered o to 7 and were of the following
diameters :—

Diameter of -

. e . . . . e e O was I inch.
55 . . D D . . . a LX IP 13 inches.
25 e L] o H D D . . . 2 وو‎ 2} 3,

» . e e . . . H . . 3 35 3% 393

e . D . D . e 4 25 5 <9‏ . . ود

وو 6 ری 5 . . LJ‏ . . . 55

?5 d 2 e e ° . e e 6 an 7 ود‎

"TI D e * D e D D 7 CT 8 وو‎

) 294 )

APPENDIX A: 6 E. GRIMES. 295

Of these, Nos. o and 1 had fallen over towards the north in directions which
were measured by the protractor as 4° E of N and 4° 30’ W of N. The only other
piece of evidence I could obtain of the direction of the earthquake was a state-
ment made by the telegraph master that the water in the tank at the back of the
telegraph office was, at the time of the earthquake, swaying from north to south.

Assam-Bengal Railway .—4At Badarpur I started the examination of the Assam-
Bengal Railway. In the section, at present unopened, north of the Barak»
between Badarpur and the hills to the north, there was very little damage,
although for part of the way it runs over recent alluvium. In the first tunnel,
however, of the hill sections some rather big slips had occurred, and along the
southern scarp of the Khasi Hills, facing the Surma Valley, large bare patches
may be seen where landslips caused by the earthquake havetorn away the jungle
from the slopes and laid bare the rock beneath.

On the unopened section of the railway between Badarpur and Silchar the
damageis very severe, especially to the bridges. Fora short distance from Badarpur
.this line runs through a number of cuttings in hills formed of tertiary beds,
and here it is but little disturbed, the chief damage being the slipping in of the
sides of the cutting. A little further on, where the line has left the hills and is on
the flat open country, mostly bhil land and all covered bv a new alluvium, the
damage is much more considerable, and for some way the bridges are all badly
damaged, especially those over the Dalasar Nadi, the Gangru Gong near
Polaspar, and the Bara Khal. In the case of the first two bridges the piers are
leaning over to the west ; probably they are not broken, but, owing to the stirring
up of the siltin which the foundations are sunk, the masonry has, with the shock,
tilted bodily over to one side in its then partially liquid bed. The embankments
closé to these piers are badly damaged and sunk in places, the damage being
caused in the same way as I have described for the road at Muktagachha.

In 'the case of the bridge over the Bara Khal the piers have fallen right over
into the river and disappeared entirely ; before the earthquake this bridge had
eleven piers standing, each with 9,000 cubic feet of masonry, there being three
spans of 60 feet and six spans of 4o feet, but after the shock only two piers on each
bank were left standing and the intermediate space is quite blank. At the time

.of the shock the two embankments are said by those who were present to have
first moved towards one another and then apart, when the telegraph wires were
snapped across and some of the insulators were hurled violently for a considerable
distance backwards from the river.

In the neighbourhood of the railway line the country isin places seen to have
sunk considerably, and near the bridge of the Gangru Gong only the tops
of several huts are now above water. The chief caseof sinking in this part of
the country is that of an area a little east of Polarpar, which extends for
some distance tothe south, and of which the sinking near the Gangru Gong
bridge is a part. Where the boundary of this sunken area crosses the Govern-
ment road, it is almost at right angles tothe road, the cracks being roughly
north and south, and the embankment is not broken down, but the crack across the

road has faulted it about 4 feet,

( 295 )

296 OLDHAM: GREAT EARTHQUAKE OF 1897.

From Badarpur to Akhaura I went along the line by trolly and ballast
train with an inspection party, the following being the chief points I noticed:
during our rapid journey, when we travelled over 128 miles of line in three
days. The damage is almost entirely confined to where the railway runs
over the flat bhil country, and where the deposits are quite recent, but wherever
the line runs through the low hills of harder and more consolidated tertiary
beds, there is little or no damage, and especially in these places we were able to
travel by ballast-train for several miles at a time. When crossing the flat
alluvial country, the embankment was in very numerous places broken down,
and although it was nearly all repaired at the time I went over it, there were still
many very bad places where the line had sunk several feet, this sinking being, as a
rule, caused by longitudinal cracking due to transverse, or obliquely trans-
verse, waves and outward movement as in the case of the road from Maimansingh
to Muktagachha. Intwoor three places where the surrounding country is very
marshy, as at mile 213 between Dakshinbagh and Kalaura, where the direction
of the alignment is S 30° W, and also at a place just south of Daragaon,
the embankment has sunk into the disturbed ground below, and has caused
a rising of the ground, or has come up itself at some little distance from the
lines at the sides, forming long, low ridges in the first case and filing up the
borrow pits in the other.

The bridges are nearly all in the flat alluvial dem and most of them are -
more or less damaged. The following is usually the nature of the damage done to
a bridge :—

The abutment walls are cracked or broken and have. come to, shortening the
span. This is the case when any of the shock acts longitudinally to the embank-
ment, and it is caused in a similar manner to the cracking of river banks, as
described by Dr. Oldham in Memoirs of the Geological Survey of India, Volume
XIX, the first semiphase of the wave on the side from which the earthquake
is advancing and the second semiphase on the other side tending to throw the two
abutment walls together; this coming to of the abutment walls is often quite
considerable, and in spans of 20 feet it is sometimes as much as a foot. Accom-
panying this movement, we in almost every case see that, on one or both sides of
the bridge, the embankment has sunk several feet. Where the bridges have wing
walls to the abutment, this forward movement has cracked and considerably
damaged them, but where, as in the case of many of the smaller bridges, there
are straight return walls, the pressure has acted along the length of the walls and
not across them, and so the bridges have mostly escaped with little or no damage.
The coming to ofthe abutments and consequent shortening of the span has either
resulted in the buckling up of the girders in the centre, or the girders have pushed
back on and broken off the balance walls of the abutments. Ina few cases the
piers of the bridges have been tilted over to the side, but in most cases only
inwards.

The following are a few notes on some of the bridges which were damaged : —

Madhob Cherra Bridge, mile 217.—The end abutment walls have sunk into
the ground and have been tilted inwards and twisted so that the face of one has

a bearing of E 33° S and of the other E 16° = the direction of the line

( 296 )

APPENDIX A: G, E, GRIMES. 207

over the bridge being N 20° E. The balance walls of the abutments have
been broken and forced back, and this, with the sinking of the abutments, has
caused the line over the bridge to be arched and the girders to be torn apart at
their junction on the centre pier, which centre pier was not affected by the
earthquake. The embankment on the north side of the bridge was badly
broken down.

Furi Bridge, mile 212.—The two centre piers have moved 18 inches bodily
towards one another and are not tilted.

Markal Bridge.—The piers on the east side of the bridge are leaning towards
the west, and the piers on the west side towards the east.

Manho River Bridge.—The line over this is in a direction of N. 15% E,
and the piers are little, if at all, damaged. The north abutment wall and
the adjoining embankment are broken down and the banks of the river on this
side are badly cracked.

Dhulat Bridge, mile 188.—The centre pier is leaning to one side and the line
onthe top is curved, and the end piers are leaning inwards. The original
direction of the bridge was N 42° E; the embankment and river bank on the
south-west side are both very badly braken down.

Khwa Bridge, mile 162, near Shaistaganj.—The direction of the line over
the bridge is nearly east and west. The abutment wall onthe eastern side is
leaning over to the east ; ithas sunk about 2 feet and its balance wall is broken
off. The next pier is, however, leaning towards the west, and the girder between
the abutment and this pier has slipped off and is now supported ona pile of
sleepers. The face of the abutment hasa dip of 84° to the west and the face of
the pier a dip of 81° to the east.

Besides the breaking down the embankment, the railway line is in many
cases bent into curves, These had almost all been straightened out again at the
time,of my visit, but some severe cases were still to be seen. The force required
to bend the line into curves must have been very great, asin many places the
steel rails are bent and curved, and the nature of the curving of the line sometimes
seems to indicate movement of the ground horizontally.

Of the horizontal movement of the ground I was, however, able to obtain a
definite piece of evidence. Between Daragaonand Shaistaganj the embankment,
which is quite low, was much broken down and the line twisted into curves for a
considerable distance, and asI was going over it on a trolley, Mr. F, P. Anderson,
the Executive Engineer, Shaistaganj Section, Assam-Bengal Railway, pointed
out to me that the line was shifted several feet from its original position. As the
alignment here was perfectly straight for a considerable distance and part of it
had not suffered in the slightest, this could be tested with certainty, and at my
request Mr. R. K. Coxe, Assistant Engineer, Shaistaganj Section, set up his
theodolite and took sights along the line. The place we chose to test was a small
bridge of three 20 feet spans in mile 164, which was in the centre of the disturbed
portion : this bridge was somewhat cracked and broken, but it was absolutely
certain that it had not moved relatively to the surrounding country and trees, yet
the theodolite showed that the centre of the bridge was 6 feet 9 inches out of the
straight line. At the time it was built it was 6 inches out and so now it is 6 feet

( 297 )

298 OLDHAM: GREAT EARTHQUAKE OF 1897.

3 inches from its original position, and as it has not moved relatively to the sur-
rounding country, it proves that the ground around, which is composed of newer
(recent) alluvium, must have shifted bodily towards the north.!

At this place and many others along the line where the railway runs over
beds of newer alluvium there are numerous cases of cracking of the ground and
outpourings of sand and water, the sand having the same characters as that
ejected at the other places I visited. At one place near Shaistaganj a native
said that petroleum came up with the sand, but I was unable to verify that
statement. Traces of sand craters very much washed down were seen in one
or two places.

Comillah.—Just before the earthquake at this place a noise was heard, which
the Collector described as the noise of trains coming from the north-east (others
say north-west), but after the shock he heard nothing. The District Engineer, how
ever, told me that the noise continued during the shock and for about 3 minutes
after, and then there were two reports like gunshots in the distance, and after
the first report the earthquake rumbling ceased ; the direction from which these
sounds came was the north-east. The information that the rumbling con-
tinued after the shock was confirmed by the reports of other people, but it
seems then to have been less loud than before. The reports were heard by several
people, both Europeanand native, and by somethey are said to resemble the Barisal
guns. These latter reports are heard faintly at other times in Comillah, chiefly
in the night between 2-0 A.M. and 3-0 A.M.

In Comillah the damage done by the earthquake is less than in Dacca, and
of the pucca houses, which I saw, only two had to be evacuated after the earth-
quake. Of the direction of the earthquake-wave the evidence is somewhat
contradictory, and in the following I will give accounts of the chief cases of
damage and evidences of the direction of the shock. The belfry of the church
was cracked across by a horizontal crack and the top part was shifted forward in
a direction of W 20° N, which is the direction in which the church is set.
The top, and three side, minarets of a Hindu shrine in the bazar had fallen, but
the pieces had beenremoved beforethe time of my visit ; the places where they
had fallen were, however, pointed out to me by the keepers of the shrine, and from
this, and the marks where the falling pieces had struck a side cornice, I put
the direction of movement as S 4o? W. The District Engineer noticed the
direction in which the waterin a small tank at Chaumata was swaying at .
the time of the earthquake, and this was almost exactly east and west. The temple
of Jaganath, about a mile outside from the bazar, was very badly cracked and
damaged, part of its eastern face having fallen out, and since the earthquake
the temple is leaning over in a direction which I measured as near as possible as
S 20° E. The nat mandira attached to the temple has its walls built exactly
north and south and east and west, and the walls on the west and south sides
have both fallen completely down outwards. :

In the tank at the back of Mr. Delayne’s house the water was set swaying” by
the earthquake in a direction roughly north-west and south-east. From the top
of the Sub-Judge's Court the western parapet wall had fallen crashing into the

oof of the verandah below.

E 1 See p. 94.
( 298 )

APPENDIX A: G. E, GRIMES, 299

In Mr. Smart’s bungalow a lamp, hanging on a wire 5 feet 103 inches long, in
the verandah, was set swinging by the shock in a direction which, as pointed
out to me, was N 35° W. One of the ladies in the station saw a circular
haystack rocking to and fro in a direction approximately N W and S E.
About two miles from the station to the west and close to one of the main
roads there is a Hindu mutt from which the top has fallen ina direction of
N 35° W. ‘This top, before the earthquake, was evidently very shaky, and
the pieces which have fallen, and arelying on the ground near the base of
the tower, are very much broken up. Nearthe road which goes from the Dak
Bungalow towards the river and close to the railway embankment there are two
mutts which are leaning overtothe south-east, the larger one. in a direction
of S 30° E and the smaller one in a direction approximately of S 25° E
From the Kaliajuri mutt the top had fallen off, and although at the time of
my visit most of the debris had been removed, the mark where the falling
piece bad struck the ground,and a little of the debris, was still to be seen, and
from these I deduced a direction of N 10% E for the earthquake-wave. In
a brickfield close to the railway station there is a mutt, from the top of
which some bricks had fallen, probably separately, and these were lying on the
ground close to the base of the tower in a direction of N 20? E from the centre
of the tower. Another mutt in a small basti on the other side of the river also
has its top fallen off. This shrine is built on the top of the bank of a tank, and
its top, which came off bodily, fell on the slopes of the bank, and then slipped
down it a little, but the marks in the ground tell where it fell, and the direction
is to the N E (N 55” E). From the following table we shall be able to see
better the evidence which we have of the direction of the earthquake-wave:—

The belfry of church , ` „indicates W 20° N and S 20? E,

Tank in Chaumata ^ : E E a لاہ‎
Hindu shrine in bazar 7 p a N40 E. , S 40° W,
Temple of Jaganath . i s N 20° W ,, 8 20° E.
Tank back of Mr. Delayne’s شس‎ 5 NW ^W.
Lamp in Mr. Smart's bungalow +. رو‎ N35 W. ,,$35 E.
Haystack : ` DEO NW » SE,

Hindu temple on road to Went E} a Has W. au S35 E,
Big mutt, Dák Bungalow Road . وو‎ MM SOE.
Small وو وو‎ » H . وو‎ N 35° W ول 5 دد‎ E.
Kaliajuri mutt . * ST و‎ » Nic E ,, Sio*W.
Mutinhrchüedd ^ . u . » تا مر اھ‎ , 57W.
Mutt beyond riwer . « « 455 N55 E » S55 W.

Looking down this list we find that the majority of these indications,
including some of the best of them, point to a direction about N 35° W and
S 35° E, and we may, I think, take this as the general direction for the earth-
quake, but the cases of the last three mutts point to some waves at least coming
from a north-easterly direction.

For the intensity of the earthquake shock and the velocity of motion of the wave
particle the only building from. which any evidence can be obtained is the mutt on
the other side of the river as, in the case of all the other mutts, the top of the fracture
is very irregular, and the fallen pieces may possibly have come down separate. This

mutt, which, as before described, is built on the top of the sloping side of a tank, is

( 299 )

300 OLDHAM: GREAT EARTHQUAKE OF 1897.

polygonal in its lower part, and on this is built a tapering spire, the top of which has
fallen in one piece on to the bank of the tank. After falling the top has slipped a
little, but the mark in the ground shows where it first struck it, and also shows that
the top was at least partially overturned before it reached the ground, where it is
now lying on its side with its base towards the tower. The following necessary
dimensions and measurements were made for me by the District Engineer ;—

Length of fallen piece e . e . e e 6 feet.
Diameter at base وو‎ . . . ۰ 3 e 2 feet 6 inches.
Do. at tap, 55 5 e ° . 8inches.

Horizontal distance of fallen piece from the centre of
the mutt, z.e., horizontal. distance of inner edge of
mark where top struck the ground from the centre

of the mutt . . . . . . ; » 12 feet,
Diameter of tower at its base . . . . e 12 feet.
Height of base of mutt above where the top first struck

ground . e e e S e . A . 3 feet.
Height of mutt as it now stands . . . . e 43 feet 6 inches.

From these measurements, I think, a minimum value for the velocity of the wave
particle may be calculated.

Mogra.—Between Mogra and Akhaura about 30 miles north of Comillah, I visited
a Hindu temple which had fallen completely over, only small pieces of the bottom
of the walls being ¡eft standing; unfortunately, at the time I saw it, the greater part
of the debris had been removed by order of the Maharajah of Hill Tipperah, in
order that the idol might be rescued, but there was still sufficient evidence remain-
ing for me to measure the direction in which the temple had fallen over, and my
reading was S 30° E, which would indicate a direction of the earthquake-wave
very nearly the same as that I obtained for Comillah. One of the natives here
told me he heard a noise, like a sharp clap of thunder in the distance, about 15
minutes after the earthquake.

A minaret had fallen freely from one of the Hindu shrines near Mogra station
in a direction of N 30° W, and the south-east corner of the pucca wall at Mogra
Headquarters House had fallen out. |

Chittagong.—The earthquake was not felt so severely at Chittagong as at the
other stations I visited, and the damage there was confined to cracks in few houses
of which the following are the chief cases. In the Anglican Church there are several
cracks both north and south and east and west, but the former are most prominent,
one of them running between the chancel and the nave. In the Old Municipal
Hospital there are several bad east and west cracks, one of which runs along
the verandah floor in a direction W 10° N. Right across the Magazine
at the Police Lines there is a big east and west crack, and there are
also some north and south cracks, but not nearly so bad asthe east and west
one ; the walls of the Magazine are built in directions only 3° from due north and
south or east and west. Right across the Municipal School there were large
cracks, the most promiment of which had a direction of E 20? N. The Sudder
Ghat Police beat-house was very badly damaged, and cracks parallel to the
walls, which are built exactly north and south, have practically divided the house
into three parts. These cracks are both of them worse at the southern end of
the house than at the northern.

Kee

APPENDIX A: G. E, GRIMES. 301

One of the clerks in the District Engineer's office told me that directly he felt
the earthquake he rushed out of his house to watch the water in the tank close by.
The size of the tank is 80 feet by 85 feet, and, according to him, the direction
in which the water was swaying, when he first saw it, measured as W 20° S,
and he tells me that this gradually turned in a direction opposite to that of the
hands of a watch until the water was moving almost exactly north and south.

Sirajganj.—At Sirajganj the earthquake was very severely felt, the Sirajganj
Jute Mill being badly wrecked, and many of the houses in the bazar in
complete ruin. Unfortunately, although so much damage has been done, there
is nothing to indicate the direction from which the earthquake waves came,
and without this information other observations, from which the angle of
emergence, velocity of motion of wave particle, etc., might be calculated, are of no
avail, and so I will merely givea few notes on some of the most prominent
buildings damaged by the earthquake.

The Sirajganj Jute Mill had suffered very severely, so much so that the
building has had to be abandoned ; all the walls are very badly cracked and some
have fallen down, whilst others are leaning outwards, the roof has fallen in
places, and the top of the chimney has fallen down. The damage, too, is increased
by a number of large cracks in the ground underneath the mill and across its
compound, which are nearly all running in a direction about N 55? E, and those I
measured varied in direction from N 38? E to N 70° E, but the biggest and most
prominent ones are in the direction mentioned above, which is about a mean
between these last two ; the damage effected by these cracks is increased, too, by
the sinking of the ground on one side ofthem. صا‎ the walls of the mill buildings
there are not so many windows, doors, and other openings as in private houses,
and the following is a list of some of the cracks, whose dip I measured :—

In wall bearing N. 26? E, crack dips 57? to S.

» دو‎ E 12^ 5 3) 51? to W.
دو‎ » E 20° S, وو‎ 46° to W,
دو‎ 27 E 80 N, وو‎ 48° to W.

From the mill chimney the top has fallen; before the shock it was 110
feet high, but at present is only 79 feet high, 31 feet of it having fallen. This
fallen part did not fall er masse, but the upper portion of the stack apparently
crumbled with the violent agitation, and the bricks fell separately, or a few at a time,
and the upper surface of the part left standing is very irregular. The bricks were
lying all round, and an examination of the debris showed that there had been very
little cohesion between the bricks and the mortar, as almost all the bricks after the
fall were clean, or else the mortar could be separated from them by hand; the
mortar was very inferior, with very little lime in it, and at the time the chimney was
built the bricks had evidently not been properly steeped in water.

The southern walls of a row of godowns have partly fallen out towards the
south (Plate XXIX, fig. 1). The walls bear E 19° S, and are 2 feet 1 inch
thick, and the side walls of the building to which they are joined are 15 inches
thick. In the worst case the upper portion of the wall has fallen outwards after
fracturing the junction with the end walls, and an examination of the debris shows
that the upper part of the wall has fallen bodily. The roof is made of galvanised

("301")

302 OLDHAM: GREAT EARTHQUAKE OF 1897,

iron, which is supported on girders, the main ones of which rest on cast-iron
pillars let into the side walls, and as only the light cross girders rest onthe end
walls, which have fallen, there was very little pressure from the roof on them.

The manager’s house was badly cracked, especially at one end (Plate
XXIX, fig. 2); here in a wall bearing N 16° E, there are cracks with
dips of 53° and 54% In the bazar almost all the pucca houses were wrecked,
and, after examining a number of them, I came to the conclusion that there was
no evidence of the direction of the shock, as, if in one house the N W and S E
corners were the worst, in another the N E and S W corners were so. Half
the post office has fallen over to the south-west, but this is probably largely
due to there teing a tank close to the house on this side, the banks of which are
very badly cracked. In the jail the upper part of the wall in the south-east
corner and part of the south and east walls have fallen outwards. The southern
wali bears W 5° N, and the upper part of this for a length of ہو‎ feet
had fallen: the height of the wall before the shock was 14 feet 6 inches, and its
thickness 20 inches, and the lower part left standing afterwards was 6 feet 6 inches
high. The eastern wall was at right angles to the southern, and of the same
height and thickness as it, and also the lower part left was of the same Bee: as
in the other case; of this wall a length of 71 feet had fallen.

Both in Sirajganj andinthe surrounding country the ground is very badly
fissured, but the direction of these cracks varies very greatly. Although there is
a tendency for the Dan Bandi River banks to be fissured parallel to the stream,
these cracks are quite subordinate, and the main fissures run obliquely to the river,
and even cross it, still keeping their direction. As at every other place I have
visited, which is onthe newer alluvium, sand and water was forced up through
the fissures, and has spread over the surrounding country, and in appearance the
deposited sand is exactly like that seen at other places. I was told by those who
watched it that the liquid mud continued to be forced up for about half an hour
after the earthquake ceased, probably owing to the weight of the fissured surface
beds pressing on the semi-liquid stratum below. Of the depth from which the
sand came there is no visible evidence, but I am told that when digging wells
at Sirajganj, they find first 25 feet of mud and then sand like that forced up.

-Tanks and other excavations were points of weakness in tre upper bed of mud,
and into many of them the sand and water has been forced up, and in some cases
the tanks are entirely silted up.

Within half an hour of the earthquake the Dan Bandi River at Sirajganj
had risen 4 feet, but this rise went down during the next two days. -

(4) Report by Mr. E. VREDENBURG, Assistant Superintendent, Geological
Survey of India.

WESTERN BENGAL.

Bhagalpur.— The intensity of the earthquake, so far as can be judged by the
damage done to buildings, was about the same at Bhagalpur as it was at
Calcutta, possibly a shade less.

( 302 )

APPENDIX A: E, VREDENBURG, 303

The buildings principally damaged are the jail and the church.

In the jail the most serious damage has been done to the factory chimney, a
slightly tapering structure built on a square plan and oriented N 62° E, N 28° W.

The sketch, fig. 32, represents the appearance of the side facing S 62° W.

The upper part of the chimney is uninjured ; the
side facing S 28° E also shows no signs of injury,
while the opposite side facing N 28° W has four
horizontal cracks, in which the lower part has
moved outwards away from the upper portions,
forming a series of steps and giving rise to the
inclined cracks visible on the sketch. They are
quite similar on the side facing N 62° E, but
slightly wider.
The part of the structure which has been
displaced has moved towards a direction N 28% W,
or perhaps a few degrees more west, on account
of the slightly wicer cracks on the north-eastern
i face. This may not be the exact direction of the
shock, since the structure is a square one, and
moves more easily along certain directions. There-
fore the direction may have been several degrees
further north or further west than that along which
the masonry has moved. As other evidence points more nearly to an east-to-west
direction, we may abandon the supposition of its having been more towards the
north than the direction observed. It is, therefore, more probable that the direction
of the shock was some degrees more to the west than that in which the masonry has
been observed to move, the limit, however, being the S 73° E—N 73° W diagonal,
for had the shock followed exactly the direction of the diagonal, there would
remain little reason for the difference in the manner in which the several faces
have been affected.

The District Engineer in his reports made out the direction to be from N 10° W
towards S 10° E. The reasons for this estimate are not given, but jtaking
into consideration the facts above referred to, the direction given appears to be
too near north and south; weseealso inthe case of the chimney that the
portion which is displaced has moved towards the west and north, not towards
the east and south. The structure must have been swinging as upon a hinge
round various layers of bricks of the wall facing S 28° E, the oscillation of
the chimney being in a direction S 28° E—N 28° W. This must have
loosened the lower part of the structure and allowed it to move to the north-west
under the impulse of the shock. As previously observed, however, it is probable
that the direction of oscillation was influenced by the shape of the structure, and
we may take it as a rough estimate that the direction of the shock may have been
about E 30? S towards N 30° W.

There are some slight injuries in the prison wards. These are long red-brick
buildings, consisting of a ground floor and upper story with a few transverse
partition walls, disposed in a radiating manner round acentral tower. This tower
is intact. The only injuries in the wards are in the transverse walls on the upper

( ape )

Fig. 32. Chimney in jail
at Bhagalpur.

304 OLDHAM: GREAT EARTHQUAKE OF 1897.

floor, and they are so evidently the result of peculiarities in the construction
that they are not very useful for determining the direction of the shock,

One more building is badly injured in the jail;
but it is designed in a manner specially liable to
suffer from a fatality of this kind. It is a small
watch-tower, a polygonal building with arches
supporting a dome. The key of each of these
arches has become displaced by the shock, and
determined at once a gaping crack which could no
longer close up. These have given way inthe
Same manner on all sides, the dome itself has
remained without support and is cracked. This
building, however, affords no clue as to the disc Wu a
direction of the shock; it will be seen from wall, Bhagalpur.
the plan of the building that the arches are
particularly weak as there 1s nothing to bear the horizontal pressure resulting
from them, and as soon asone of the voussoirs gives way the structure is hopelessl y
wrecked, for the cracks cannot close up again as they might in a building con-
sisting of horizontal layers.

The choir of the station church has been ruined by the fall of a pillar and of
the roof which it supported. It is difficult to say whether it is the pillar that
brought down the roof or the roof that brought down the pillar. It would seem
that the damage is to be ascribed rather to peculiarities in the building than to
the violence of the shock. | ;

The building consists of a central nave with two aisles. As will be seen on
the following diagram, the buttresses give the building an illusory appearance
of solidity for they merely strengthen the outer wall, which is quite disconnected
from the inner row of pillars and arches which have to bear the weight of the.
central roof. This roof is supported by beams and by wooden arches, both
transverse and diagonal. The beams are all transverse and are placed both upon
the pillars and the centre of the longitudinal arches of the masonry wall.

بے
2
IO‏ .=

Fig. 34. Cross section of church at Bhagalpur.

( 304)

APPENDIX A: E, VREDENBURG. 395

The wooden arches can only bear a fraction of the weight of the roof, the effect
being quite different from that of a stone roof
with ordinary cross-vaulting, which carrie sthe
entire weight upon the pillars. The transverse
beams carry a considerable proportion of the
weight onto the masonry wall at its weakest
points, being the points of the lateral arches,
each of which carries a beam on either side as
shown in the section. The effect differs also
from that of an ordinary ceiling which would Fig, 35. Plan of roof trusses in
have many more beams, distributing the church, Bhagalpur.
weight far more evenly.

Another cause of weakness is the small distance to which the beams penetrate
into the masonry, which appears out of proportion with their span, The most
probable explanation, therefore, is that one of the beams was actually dislodged,
and, in falling, carried down the roof, which in turn wrecked the pillar. It is the
eastern pillar at the extremity of the northern series of the inner walls. The
building is oriented exactly east and west, the choir occupying the eastern end.

I have gone into these details at some length in order to show that the damage
to this building is the result of somewhat special circumstances; otherwise it
might give an erroneous idea as to the intensity of the shock at Bhagalpur. The
remainder of the building is almost without any injury. There is a crack through
the window in the east wall, but this might be merely the result of the falling of the
adjoining pillar and roof.

The little pinnacles and turrets are all uninjured with the exception of those a
the west end which have had the plaster and bricks shaken off their upper part,
merely the knob remaining, supported on an iron rod. A small belfry springs
from the centre of the pediment; its upper part has broken but has not fallen,
owing to the iron rod that runs through it. It is bent in a north-west direction.

According to tfie station master and the telegraph clerk at the railway station,
the earthquake began at 4-25 railway time; the telegraph clerk says it lasted
till 4-29 and the station master until 4-30. I have ascertained, however, that
these railway clocks are not strictly correct; variations may amount to as much as
two minutes.

Jamalpur.—This town is built on a very regular plan, All the houses and all
the workshops of the East Indian Railway are oriented in the same manner,
about N 20° E—N 70° W.

The effect upon the buildings has been the same in most cases : the walls facing
north or south are cracked, while those facing east or west are considerably
displaced from the vertical, or have actually given way ; the effect remains similar
whether the longer axisof the building is in a north-south or an east-west
direction. |

The violence of the shock appears to have been very great; very few two-
_ storied buildings have remained intact; the corners have often collapsed through-
out the entire height of the building. In most cases the upper stories had to be
taken down at once and many houses will have to be entirely rebuilt.

x Kugener »)

306 OLDHAM; GREAT EARTHQUAKE OF 1897.

The sketch, fig. 36, gives a tolerably typical example of the kind of damage,

Fig. 36. House at Jamalpur, facing S 20° W.

though the degree of injury is here comparatively slight. The side of the house
‘shown in the sketch faces S 20° W. '

The chimneys situated upon the roofs of houses have given way without
exception, and it will be noticed that whenever the lightning conductor has been
bent by the falling structure, it is in a westerly direction. Thisis all the more
remarkable as the chimneys, which are square, have their diagonals corresponding
with the axes of the houses.

The next sketch shows a house in which the western front presents no cracks
(the north-east corner of the house has totally collapsed); the lightning conductor
is seen to be bent in exactly the same direction. l

Fi .37. House at Jamalpur, facing W 20° ۰

( 306 )

APPENDIX A: E. VREDENBURG. 307

More remarkable still are the houses situated in a symmetrical position, with
their front carrying the chimney facing east instead of west. In that case the
chimneys have fallen inward instead of outward, dragging the lightning conductor
after them, so that it is still bent in a westward direction.

Taking into account the orientation of the houses, these results agree with the
direction as indicated at Bhagalpur, that is, the shock would have proceeded
from a direction S 20° to 30° E,

Mr. Rednell, the Locomotive Inspector, happened to be outside his house at the
time the earthquake began, and he watched it very carefully from the commence-
ment; herealized the nature of the phenomenon at once owing to the vibration of a
large iron roof built over the terrace of the house. The earthquake commenced at
4-27 railway time, and lasted seven minutes. According to Mr. Rednell, the first
vibration, which was very moderate, was approximately in a northssouth
direction, and lasted about two minutes before the shock became more serious and
assumed its east-west direction.

It should be noticed that Jamalpur (like Tinpahar and Sahibganj) differs
from most of the towns I have visited, in being built at the very foot of the hills,
with rock appearing at the surface in many places, while the other towns are
built upon a great thickness of alluvium.

One of the chimneys in the workshops, built on a circular plan, was damaged,
but it had already been pulled down when I arrived ; the crack, as described to me,
was vertical and ran through the chimney in an east-west direction; the northern
side was bulged out a little more than the southern.

The workshops are oriented just like the houses, but with their longer dimension
east and west. instead of north and south, the walls facing east and west are torn
away from the building or have collapsed.

The church and one large building next to it have escaped without a single
injury. This is perhaps largely due to their excellent construction. A very short
way east of thesestwo buildings, the ground slopes suddenly towards a large
reservoir ; if we suppose the shock to have come from the east, this solution of
continuity in the surface of the ground may not have been without influence in
interrupting the vibration.

Monghyr.—The damage appears to be about the same in degree at Monghyr
asat Jamalpur. Many of the one storied houses are injured, but they are of a very
inferior style of construction compared to those of Jamalpur.

One of the worst damaged buildings is the Baptist Chapel. The plan is not
unlike that of the church at Bhagalpur, as it hasan inner room with flat ceiling
and an outer portico, these parts being without any connection other than that of
the wooden beams which they support. The walls of the inner hall, however, have
more consistence than at Bhagalpur; they have no arches, but only a series of
doors in their lower part.

The axis of the building runs north and south. Over the porch, which faces
north, is a square tower with curious round windows. This tower is cracked right
through both from north to south and east to west, the northern and southern faces
“being the most injured. A great portion of the masonry has fallen southwards,
bringing down with it the roof of the church ; in other parts of the building the roof

X gre (387)

308 OLDHAM : GREAT EARTHQUAKE OF 1897.

has not collapsed, but the walls have been moved out of the vertical, the beams
being displaced to a considerable extent. The effect is what might have been pro-
duced by a shock travelling in an approximately east-west direction. The porticoes
to the east and west are also affected iN the same manner; at the corners of the
building the arches facing north and south are much more damaged than those
facing east and west. A portion of the cornice from the eastern portico fell east.

The E-W direction of the shock is clearly shown in another injured buildiags
the clock-tower. This is a structure in red bricks rising above an arched
gateway, facing W 5° N. The eastern and western wall of the tower are
supported by flying buttresses; these have acted as a protection to the arched
windows placed between them, but as the shock came from the east, and there
is no structure of the same kind protecting the north and south walls, the lower
part of the tower, as far as the upper part of the flying-buttresses, has moved
bodily towards the west, causing the collapse of the arched windows on the north
and south walls; the damage is particularly great on the north side.

One of the buildings accompanying the little mosque in the cemetery has had
its western extremity injured by a crack inclined at about 85° to the horizon and
hading west. It shows equally well on both the northern and southern aspects.
The buildings, however, are placed on a small mound, and the outer walls, from
their situation, would be predisposed to injury.

In a house not far distant the east-west direction is better shown by a wall
facing west, which has become detached, notwithstanding two solid buttresses,

The Collector's house, a one storied structure built on an eminence, has a semi.
circular verandah facing 5° north of west, which is very badly wrecked ; the sym-
metrical one facing 5? south of east is also damaged, but to a less extent. The
rest of the house is not much damaged except for a few arches.

The post office is a one storied building, with a verandah facing west, and a
smaller one facing east. These are carried by pillars which are not built on the
same plinth as the house. Neither of them has collapsed, but both are damaged
and rent apart from the house, the western one especially. The rest of the building
has suffered little.

Many one storied houses and the tall houses of the bazar are badly damaged,
but, owirg probably to their imperfect construction, the cracks fcllow no spiecal
direction, and their ruin merely bears witness to the violence of the shock.

Purniah.— At Purniah the houses are built in such a manner as to give little
idea as to the direction of the shock. They are tall buildings made of indifferent
bricks jointed with mud or bad mortar, and coated with an enormous thickness of
plaster. They have been shattered without the formation of very definite cracks.
Comparing them with buildings of the same style at Monghyr, it seems, however,
that the violence of the shock was much greater. At the same time One of the
Government Offices, a new red-brick building, with an upper floor and a square
clock-tower, has remained intact, and the clock was not even stopped.

The ground cracked in every direction wherever it was unsupported, and jets
of water escaped from a number of these fissures. Some of them were still open
at the time of my visit, for instance, at Harda Bridge, 3 miles south of Purniah.
The road at that place is raised some ten to fifteen feet above the level of the

( 308 )

APPENDIX A: E. VREDENBURG, 309

Surrounding alluvial plain. The fissures ran parallel with the direction ef the
road along its western portion, and the western slope, asfar as the bridge, for a
distance cf 750 feet; the low ground under the bridge was flooded when I saw
it, but at the time of the earthquake it was dry and the cracks extended under
the bridge for another 150 feet. The bridge is an iron structure anu is not
injured. The road at this place runs N 30° E. When I saw it, it was
being rapidly repaired, and a gang of coclies were busy filling the fissures with
rubble; yet in many places the cracks were still more than ten feet deep; some
were over a foot in width.

It is not merely the somewhat unsupported ground of the embankment which
has been effected, for some of the fissures are in the plain at the base, so that a
permanent displacement of the ground must have taken place at the surface, at
least in the immediate neighbourhood of the embankment.

Fissures were openéd in many other places at Purniah, particularly on the
banks of rivers; they do not exhibit any constant direction, but follow lines of
weakness in the ground.

Sahibganj.—At Sahibganj, which is situated at the very foot of the Rajmahal
hills, the effects have been still worse than at Purniah. Even the low, one storied
houses of the railway officials, which are built of good masonry with iron beams,
have been badly damaged, some of them even completely ruined. When they are
examined, however, it is seen that they have given way in the weakest portions
without reference to any special direction. For instance, many houses oriented
S 10° W have had the verandah separated from the house.

The Institute is a large building oriented in the same manner; it has an upper
story, and is surrounded by a portico. Itisalmostcompletely wrecked, but arches
doors, and windows are so numerous that the cracks have merely followed the easiest
path offered to them. The building had a tower on the south side that fell south.

The railway station faces the Institute, being similarly oriented; it had also
a clock tower, placed on the north façade which fell north, causing the collapse of
the stone portico beneath.

The northern wallin a small mosque, a compact building about twenty feet
high, three sides of which at least are without large openings or with only blind
arches, has one crack hading west,but by referring to the western wall it will be
seen that the cracks follow no law, except that they are developed in the weakest
part of the building, that is, the corners. The northern crack of the main wall,
taken in connection with the crack in the northern wall would represent the
shock as proceeding from the south-east; there is a parallel crack at the northern
corner of the mihrab, butit is situated exactly in the same manner as regards
the masonry. At the southern corner of the building the cracks are in exactly the
opposite direction. The southernlittle turret onthe mihrab has fallen partly inwards
to the east, partly to the south. The southern face of the building is intact ;
the eastern face is the most injured, but it consists of arches to which the damage
is restricted.

[n the bazar, where there are many rows of two storied houses, almost every
building is wrecked.

Two clocks were stopped in the railway station; one of them stopped at 4-28

( 309 )

210 OLDHAM: GREAT EARTHQUAKE OF 1897.

railway time; the other one, in the railway telegraph office, stopped at 4-30.
Possibly they did not stop exactly at the same time; but differences amounting
to that extent do exist between the different clocks at one station, as they are
wound only once a week. The station master's appreciation of the duration of
the earthquake was that it lasted three minutes.

Fissures opened out inthe ground in many places in the immediate neighbour»
hood of Sahibganj. This is interesting, taken in connection with the absence of
cracks at Jamalpur and Monghyr; the thickness of alluvium and loose ground
cannot beless at Jamalpur than at Sahibganj, for both are situated atthe very
limit of the alluvial area, while at Monghyr the thickness must be much greater.
This confirms the conclusion arrived at from the state of the buildings,that the
shock was much more intense at Sahibganj than at the two abovementioned
localities.

Tinpahar.— At Tinpahar the railway station and smaller buildings surrounding
it are much shattered. As at Sahibganj they are in the immediate vicinity of the
hills. Therailway stationisa long building facing N 70° E, with an upper story
cccupying only the middle part of the building, as shown in fig. 38. The shaded
portions of the upper floor have entirely collapsed.

At the northern extremity, on the ground floor, is a waiting room, which shows
somewhat regular cracks; it occupies the entire width of the building. In the
eastern and western wall the cracks agree in direction ; but they are also similarly
situated with regard to the corners of the edifice. Inthe northern wall, however,

A OS‏ یی کک رسس
E ai‏
L L‏

PS

Fig. 38. Railway Station at Tinpahar.

the cracks nearly all hade eastwards; if these cracks are to be depended upon,
they would imply that the shock proceeded from a direction a few. degrees west of
south. It is not possible, however, to place great reliance in these cracks; the cracks
;n the end wall hade towards the lower ground opposite the eastern fagade of the
pedir. ; and in a small building close by, placed on low ground, and similarly
oriented, the northern wall has cracks hading in both directions.

Rajmahal.—The shock appears to have been as intense at Rajmahal ; but there
are few buildings to bear witness to it; there are a few very squat bungalows
which are thatch roofed, and the bazar consists principally of mud huts, As to
the older buildings they were greatly ruined before the earthquake,

The injuries observed do not afford much indication as tothe point of DUE
ofthe shock. The jail has a few cracks in its outer walls.

( 330 -)

APPENDIX A: E. VREDENBURG 311

A large mosque, now transformed into a hospital, has one crack in its northern.
wall, and one in its southern wall running through the apex ofthe arches. The
rest of the building and the domes are uninjured. |

The large railway station, a long building facing SE is almost entirely
wrecked owing to the walls having swayed south-east and north-west, which
has displaced the iron beams. One of the clocks in the railway station, facing
south-east, stopped at 4-30; anotherfacing north-west, did not stop. According
to the station master, the shock lasted three minutes.

At the time of the earthquake Mr. McGavin, the Collector and Magistrate,
was on the opposite side of the river to that where Rajmahal is situated. The
gtound cracked in many directions, and water and mud were ejected from the
fissures. Some of the indigo fields were entirely flooded with a kind of soft mud.
In some places the jets of water and mud formed small conical mounds with a
crater, about one foot high. The largest fissure observed was three feet wide.

The earthquake had a curious effect upon the fish in the river near Rajmahal ;
they rushed into side creeks, several of which communicated with tanks, where they
collected in large numbers.

Berhampur.—The four towns of Berhampur, Murshidabad, Baluchar, and
Azimganj are the most damaged that I have seen, scarcely a building being left
entirely standing, and even those which have not collapsed are badly injured.

At Berhampur the Government Offices are situated in large buildings, of ex.
cellent construction, formerly used as barracks. They are double-storied buildings,
facing S 20° W, The corners have generally given way ; on the upper floor part
of the ceiling has fallen down in each of these barracks, also carrying down the
ceilings of the lower story. In one of the buildings some bricks from the south-
east corner were projected a long distance, over one hundred feet, to the east, but
as the corner fell throughout the entire height of the building, which is 50 to 60 feet,
they may have been propelled rather by the falling mass than by the shock. The
greatest part of the masonry fell outwards at a short distance.

The dak bungalow, church, post office, etc., are situated in long rows of low
buildings with arches on either side. They are very massive buildings, about 20
feet in height. Both the inner walls and outer arches are considerably thrust out-
wards by the beamsthey support, which have been displaced as much as one foot
in some places. The orientation is the same as that of the barracks, S 20° W.

The southern end of one large double-storied house facing S 70° E has
entirely collapsed. In this case, however, it gave way along the cracks of the 1885
earthquake,

The tower of the college is cracked along the four corners; it is oriented N 22°
E and N 68° W. The damage appears to be somewhat greater along the north-
south diagonal. The clock in this tower has two dials; one of them is stopped at
5-10, the other at 4-45; they may be looked upon as equally unreliable. Mr.
Leavinge, the Collector and Magistrate, says that his clocks stopped at 4-55;
Calcutta time. Of course, this is only approximate. At the Telegraph Office they
were not ableto give any definite information.

The ground was fissured in many directions.

Murshidabad.—At Murshidabad most of the buildings are oriented exactly,

aan”)

312 OLDHAM: GREAT EARTHQUAKE OF 1897.

north-south and east-west. The buildings, whose longer axis runs north and south
are generally more injured than those whose
CC) principal direction is east-west.

Ter 27ج‎ E In the Nawab's new palace nothing has

Sr Neh he ae
PS mm E been left standing but the outer walls. This
ze pit

building, which is made of bricks, faces the

west, and has an upper story ; there were no

wooden or iron beams: the ceilings consisted
of shallow arches.

As these were held in their place merely by

Fig 39. Flooring in Nawab's the resistance of the outer vertical walls with-

Palace, Murshidabad. out any buttresses to oppose this exceptional

stress, they must have given way easily.

In another building of this same palace a clock-tower, facing west, fell west by
a little south.

In the stables, which were large and well built, most of the ceilings have
collapsed. The building was a large quadrangle, the north-east side of which
consisted of a double-storied habitation. The outer wall of this habitation facing `
N 40° E collapsed entirely.

The only building which is comparatively little damaged is the old palace. It
is a very large three-storied building with its longest dimension in an east-west
direction. It is very massive and well built,, largely of Chunar sandstone. The
damage is that caused by the rocking to and fro of the walls. The shorter walls,
those that run north and south, are not damaged, except for some slight cracks near
the upper corners. The longer walls running east and west are not cracked either,
but have been more or less displaced out of the vertical; the porticces and pediments
have especially been detached from the building, but none of them have fallen.

In the Durbar hall, a large circular room covered with a dome, the masonry of
which has remained quite intact, there are recesses, in which two large candelabra
were situated. They were firmly secured in their place by wires. The cande-
labrum placed in the north-west recess fell south-east, breaking all three wires.
The one in the south-east recess vibrated sufficiently to break the south-west wire,
but it eventually remained standing. Unfortunately this hall is on the first floor of
the building so that little information is given by these directions. In some other
rooms on the same floor, marble pillars 3 feet high supporting statuettes or
vases have been thrown down ; at the time I visited the building they were still
in the position in which they fell, and were found lying in all directions.

Opposite the palace there is an isolated clock tower which has not suffered.

Close by is a little Mohammedan shrine, a square building with a dome in the
centre and minarets at the corners. Like the other buildings it is oriented exactly
north to south and east to west, The dome and the walls are not cracked, but the
our minarets have fallen ; three of them fell exactly to the west ; the one placed on
the north-east corner fell due east.

I could not detect any instances giving an exact measure of the intensity of the
shock or of its direction ; but the general circumstance of greater damage to build-

(32)

APPENDIX A: E. VREDENBURG. 313

ings with a longer north to south axis would show that the shock followed an east
and west direction.

Azimganj.—Azimganj is in the same state of ruin as the neighbouring cities»
The tall and well built houses have almost entirely collapsed ; the very few which
have not fallen are much damaged. It is the only place I have visited where
temples of the Hindu style of architecture have been damaged. Those of Azimganj
are all Jain temples. One of them, a very beautiful building in Jaipur marble,
is almost totally wrecked ; the others have suffered slight injuries.

The wrecked temple consists of two courts surrounded by porticoes: it is
oriented east and west. The inneror western court is on a raised platform connected
with the outer court by a flight of steps. In the centre was a group of five towers ;
the central one fell to the east, causing the ruin of a large portion of the building.

In the three localties described above, just asin Calcutta, the-shock was not at
all instantaneous, and moderately severe vibrations lasted for a minute or two
before the buildings began to fall. This accounts for the comparatively small
number of fatalities.

Nalhati.— At Nalhati the shock was not so severe; no fissures were observed in
the ground in the neighbourhood.

Bardwan.—At Bardwan it was still less intense. The damage done is much
less than in Calcutta. I saw very few houses that were even cracked to any marked
extent. At the palace, on the western side of a three storied building, a small
portion of the balustrade from the roof fell to the west. Much of the balustrade
is missing, but on information I found it was removed since the shock as it was
considered unsafe.

Summary of effects on buildings.— The structures which have suffered most of all
are imperfectly supported arches ; the lower the curvature, the worse they have
fared, as in theinstance ofthe new palace at Murshidabad. In the clock tower
at Monghyr a very slight displacement of the western wall has caused the com-
plete wreckage of the small arches on the north and south faces; while those on
the two other faces which are supported by flying buttresses are intact.

A form of structure still worse calculated to resist this form of casualty is that of

the sentry-tower at Bhagalpur, which, being of a polygonal shape, leaves the arches
particularly unsupported.

Fig. 40. Elevation of buildings at Berhampur.

On the other hand, rows of arches like those in some of the public buildings at
Berhampur (Fig. 40) have stood very well, whatever position they may have oct-
cupied with respect to the direction of the shock, and notwithstanding that the
transverse beams they support have often displaced them from the vertical to a

(012134 ]

314 OLDHAM; GREAT EARTHQUAKE OF 1897.

considerable degree. They have laterally afforded mutual support to one another
and only the last one at either end of the series is injured.

The domes of the mosques are generally without a crack, even though the
structures supporting them may be somewhat damaged.

Buildings with iron beams have suffered less than those with wooden beams.
In one of the barracks at Berhampur, where the upper part of wall has given
way, the beams are still resting on one of the lowerlayers of masonry (Fig. (41).
A similar effect is shown in one of the photographs from Jamalpur. Whenever
this happens, it may help to save the lower stories and it is more likely to occur
with iron than with wooden beams.

Fig 41. Injury to roofs at Berhampur.

It is not easy to give exactly the relative intensity of the shock at the various
places observed. The following list, by order of increasing intensity, is however

approximately exact :—

Great damage to buildings,

but ground not fissured. Calcutta Bhagalpur.

Jamalpur Monghyr.
E

f Cauta.

Buildings totally wrecked

and ground fissured, Sahibganj Tinpahar, Rajmahal.

Berhampur Murshidabad, Azimganj.

s. Report by Mr. P. N. Bose, Deputy Superintendent, Geological Survey of
India.

EASTERN BENGAL.

My tour lay through a part of the country— Jessor, Khulna, Barisal, Noae
khali and Faridpur—which has been but little affected by the earthquake of the
12th of June. The greater portion of the area is composed of deep, soft, recently
formed, more or less homogeneous loam, and is intersected by numerous rivers
and Khals. There are no places which can claim any great antiquity; and
some, like Noakhali, have sprung up only within the last century.

Hous:s are as a rule very lightly built, having mat walls and thatched or
corrugated iron roofing. Brick built houses or temples are few and far between,
and they have usually escaped with comparatively insignificant cracks (mostly

(131449

APPENDIX A: P. N. BOSE. 315

above arches). Only one case of earth-fissure was reported to me; but at the
time of my visit it was under water, and could not therefore be seen.*

Calcutta.— Before leaving Calcutta 1 made a few observations which may be
conveniently recorded here.

At No. 5, Bankshall Street (an old two storied house), a portion of the parapet
at the north-eastern corner of the house was thrown down in direction N 20? E.

In the next house (No. 6, a very old two storied house), the walls bear N 30° E
and E 30° S. The former walls (those bearing N 30° E) were most severely cracked,
the cracks being nearly vertical (except those above arches). The walls at the
northern and southern ends of the house were slightly fissured off. The house
would appear to have swayed not very far from the direction of the walls N 30° E
—S 30° W.

At No. 2, Old Court House Street, one set of walls bears NE—SW. A
portion of the parapet from the south-western corner of the house was thrown
in direction S W. 3

Gobardanga.—2nd Fuly.—In the morning examined the zamindar's house, the
largest and most imposing building seenlin the course of my tour. Ithas been badly
cracked. The cracks, I noticed, however, were all on arches. The most interesting
fact in connnection with this house was the fall of a roundish turret-like structure
which surmounted the parapet on its southern side. Unfortunately the débris
had been removed before my visit. But from the evidence of eye-witnesses I
ascertained the direction in which the turret fell to have been S 25° W.

Reached Jessor in the afternoon. Saw the District Magistrate, the Civil
Surgeon, etc. Their impression was that the shock was in a north-south direction.

Ramnagar.— 37d Fuly.— Visited Ramnagar, about 5 miles from Jessor. The
garden house of the zamindar of that place has been somewhat severely damaged.
The cracks were all found to be on arches. The wall of the verandah bearing
N 20° E—S 20° W has been shifted northward about 2 inches. The shock here
would appear to have been very nearly in that direction.

Jessor.—At Jessor the damage to buildings was slight. Theroof of a room
in the Collector’s Court has been rather badly cracked, the cracks being parallel
and running east-west.

4th Fuly.—Left for Khulna in the morning. All the way from Jessor to
Khulna did not see a single brick built house, except the railway stations, which are
new, small, and have mostly corrugated iron roofing. They have suffered but little,

Khulna.—Reached Khulna at 11-30. Buildings here, belonging mostly to
Government, are generally new, low and well built, and have suffered but little
damage. The house occupied by the District Magistrate, the oldest in the station,
has been damaged to some extent. The parapet on the northern side has got
a long crack running nearly E—W. There are other cracks on the roof more
or less parallel to it.

Barisal.—6th Fuly.—In the Jail some walls bearing N—S have been cracked

above arches. So far as I saw, the walls in the opposite direction have not had
any cracks.

The fissure, reported to have been formed ata village called Beliya Bhekutiya some‏ ٭

4 miles west of Jessor. It was said to be about 24 feet deep, and sand and water were
affirmed to have issued from it.

و وا

316 OLDHAM: GREAT EARTHQUAKE OF 1897.

In the Reserved Police Barrack the walls bearing N 10”? E—S tro? W have
been cracked much more seriously than those in the opposite direction.

In the Circuit Houses I noticed numerous cracks on arches in N —S walls; but
only one arch on an E—W wall was found to be cracked, and that slightly.

In the office of the District Superintendent of Police arches in N—S walls
have been cracked. But in the adjoining Collector's Court, a very old two storied
building, the arches cracked are mostly in E—W walls.

Noakhali.—7:h Fuly.—Reached Noakhali about 6 P.M. Heard the “ Barisal
guns ” inthe evening distinctly. They appeared to come from south-eastern direction.

8th Fuly.—The house at present occupied by the Judge and the District
Superintendent of Police (a very old two storied house) has got several vertical
cracks on N—S walls in the upper story. The arches have been cracked in N—S
as well as E—W walls.

The Circuit House (a portion of which is occupied by the District Magis:
trate) has got a. few nearly vertical cracks in N—S walls. The earthquake here
travelled very nearly in N—S direction.

The house occupied by the Judge at the time of the earthquake was con-
demned by the District Engineer three years ago. It is a very rickety old two
storied structure. The cracks here are numerous, and are indifferently in N—S
and E—W walls, nearly always above arches. Beyond a few pieces of plaster
which have come down inside one of the rooms nothing has been overthrown.

Heard the * Barisal guns ” again this evening at intervals of 4 to 15 minutes.
They appeared to come from a southern direction. In this connection, 1 may
state that at Barisal itself the * guns," as I was informed by the District Magis-
trate and others, are not heard distinctly or so frequently as they used to be, so
I heard them there only once, and that too faintly.

Barisal.—zoth and 11th Fuly.—A clock in the house of the District Judge,
which stood on a bracket and had its pendulum swinging W 34? N—E 34° S, stopped
during the shock.

In the house of Mr. N. Gupta (barrister-at-law), a clock standing on a bracket
and having its pendulum swinging N 15? E—S 15° W, stopped during the shock.
Another clock in the same house fixed to a wall bearing N 15? E—S 15? W. did not
stop, showing that the wall could not have moved very far from that direction. .
A clock in the post office fixed toa N—S wall did not stop, indicating that the
shock may have been in that direction.

The clock on the church tower which faces south and has its pendulum swing-
ing E— W, stopped during the earthquake, showing that its motion could not have
been in that direction.

The evidence afforded by the last named three clocks—that of the first two
not being reliable—combined with that of the cracks in the Jail, the Circuit House,
the Reserved Police Barrack, etc., mentioned above shows that the direction of
wave-path at Barisal may be taken approximately to be N 15? E.—S 15° W.

Mr. D'Silva, resident of Barisal, saw the water in his tank, which has its length
N —S, move lengthwise during the shock and then take a twist and move E—W.

The Telegrapb Master of Barisal informed me that the time of commencement
of the shock at Barisal was 5-9 P.M, (local time), that is, 5-1 P.M. Calcutta time.

( 316 )

APPENDIX A: R. D. OLDHAM. 317

Faridpur.—14th and 15th Fuly.—The general impression here is that the
houses swayed in N—S direction. The District Magistrate informed me that the
District Superintendent of Police had seen the water in a tank move up and down
in that direction.

Fudge’s Courthouse.—Walls bear N 22° W and E22? N. Arches have been
cracked indifferently, but slightly, in both these directions.

Fudge’s House.—A very old house. A portion of the parapet on the southern
side of the house has come down in a SSW direction.

The Judge informed me that some bottles which were standing on a rack fell
down. From his evidence, and that of small fragments of the bottles, 1 made out
the direction of fall to have been N 29? E.

The Zilla Schoolhouse.— An arch bearing N 20° W—S 20? W has been seriously
cracked at two places. 1

Sub-Registry Office.—Waalls bearing N 10° W have got a large crack at either
end of the house. Arches in E 10° N walls have been slightly cracked.

Jail.—Two parallel buildings (wards Nos. 4 and 4) have been rather badly
cracked. The walls bear N 15° W and E 15? N. The cracks are worst on the
former walls.

The Postmaster informed me that the shock was over at Faridpur at 5-10 P.M.
local time (5-4 P.M. Calcutta time). The clock is regulated every day.

Further details from notes by Mr. R. D. OLDHAM, Superintendent, Geological
Survey of India. 1
To the foregoing I may add the following details from my own observations :
Dhubri.— The wall of the Gauripur zamindar's compound was overthrown for

the most part, but about one-third of its length left standing. Height 3! 11,”
thickness 2' 9".

Two gate pillars of a house were overthrown, one to N 5? W, the other to
S 35? E. The original dimensions were about 7’ high x 1' 6" square.

Goalpara.— Municipal boundary pillars 1' 6" square, 5' 2" high, with a trun-
cated pyramid on top 1' 4" high and tapering to 4" at top ; total height 6' 6”, All of
these were overthrown, except the one nearest the river bank.

No. 1 to N 50° E No. 3 to N 63° E
No. 2 be N 87° W No. 4 Es N 65° W
No. 5 to S 55 W

Gauhati.—One of the Gate pillars of the Loki Rani’s house fell to N 60° E: the
other had been removed.

A pair of gate pillars in the compound east of the Telegraph Club were much
broken up and overthrown. The west pillar to S 13° E, the east to S 40? E.

In the cemetery a marble cross 2' 11" high, 1' 7" across arms, 1' 11” to bottom
of arms, section 41" x 3’was broken across at the socket, which was 33” x 13”.

Another cross 3' 6" high, 2' across arms, section 53" x 32” with an oval of
11 6" x 1' 3" filling the cross, was broken at the socket, which measured 5" x 23".

The monument to Robert Beecher is a stone pillar and vase standing on a
pedestal. This had been projected and the base of the pillar struck the steps at
the foot of the pedestal in a direction S 15° E, the horizontal distance being 2' 9”,
vertical ai 9".

(917 .]

318 OLDHAM: GREAT EARTHQUAKE OF 1897.

Tezpur.—A small مسا‎ enclosure, with minarets at the corners, octagonal
section, 8" side, about 3' high was not injured. Brickwork evidently poor.

The Dak Bungalow gate pillars, 8۷ high x 2! 6” pete, were uninjured.

Dacca.—The pillars round the racecourse," 5’ high, 1! 7” diameter, circular
section, have not been injured.

Chhatak.—The tomb of Lt. H. T. Bird stands in a separate enclosure by
itself ; the gate pillars at the south side have been overthrown, the east pillar to N
the west toS. They were attached to walls which controlled their fall. Height
6' 6”, base 2’ square,

Cherrapunji.—Gate pillars on eastern road oriented N 25% W, 4' high x 2' 2”
square. Northern fell to W 13° S, southern to N 40? E.

Dak Bungalow western entrance, east pillar fell S 30° W, west وت‎ 53 E,
They were too much broken up to obtain the original orientation.

A Khasia monument in the compound, a stone slab A ei high and about
2 6" broad by 6" thick, longer axis N 2° E, has been broken across at ground
level and thrown to east.

The tallest pillar of the old aqueduct leading to Inglis’ bungalow is 7’ 9”
high and 2’ square. It rises from the bed of a small nala or ravine, and is frac-
tured and twisted, but not overthrown.

The gate pillars of Hudson's bungalow are g' high x 3’ square. Those of east
gatewaybear E 139 N ; the east pillar has fallen to S 50° W, the west pillar to
S 55° W. The western entrance bears N 30° W ; the south pillar has fallen to
W 11° S, but the masonry is much crumbled at the lower part ; the northern pillar
is intact. ,

A small bungalow near by had a number of round pillars in front, evidently
old verandah pillars. These fell to N 23° E, N 55° E, S 27° E, N 75° E.

About two miles from Cherrapunji the road to Shillong crosses a saddle, and
here an old Khasia monument in the form of an octagon of solid rubble masonry
about ol side has been overthrown and hurled to E 5° S. The debris extends for
nearly 50 feet ; the original height could not be ascertained.

Shillong.—In the cemetery a headstone over the grave. of Major Stewart was
fractured. The headstone is of the usual type with vertical sides rounded off ina
semicircle at the top. The dimensions are 2’ 13” high, 2’ 1" broad at the base and
Ge thick. It is broken across near the bottom, the fracture running irregularly from -

" above ground at one side to 3” above ground at the other. The head stone is
SC of firm sandstone, a partially weathered form of the Shillong quartzite.

A short way beyond the bridge, five miles out of Shillong, on the road to
Cherrapunji, a small Khasia monument was fractured. It is roughly 5! A high
and broken off 10” above ground, the part broken off being 4'6"long and the
fractured surface about 1' 8'by 8". "This is the most striking case of fracture I
saw asthe rock was a hard glassy quartzite and showed no signs of weathering
except for a thin layer on the outside.

Silchar.—According to Dr. Williams, one of the gate pillars of the Mission
bungalow was thrown due north. The pillar was 15' high x 2'4" square and
was slightly inclined northwards before the earthquake. Its fellow, which was
upright, was unaffected.

( 318 )

APPENDIX B. | 319

Appendix B.

Details of fissures, sand-vents and allied phenomena.
Contains extracts from official reports and replies to letters of
enquiry. The extracts given in Chapter II, pp. 14, 15—23, 25-—27,
should be read in connection with these.

Letter No. 998G., dated Rangpur, the 13th September 1897, from J. D. Cargill,
Esqr., Offg. Collector of Rangpur, to the Commissioner of the Rajshahi
Division. i
In reply to your circular memorandum No. 706 Mcl., dated the 23rd ultimo,

calling fora report on the earthquake of 12th June last, giving information as

tothe extent to which fissures in the earth, outpourings of sands, etc., have been
observed, I have the honour to report as follows :—

2. The soil of the District of Rangpur is physically divided into two
distinct divisions, locally called khiar and pali. Khiar is hard red clay, and com-
prises about 10 percent of the area of the district. Pali is alluvial deposit, con-
sisting mostly of fine sand and containing vegetable mould. In the khzar section,
which extends over the south-western part of the district and comprises parts of
the jurisdictions of thanas Govindganj, Pirganj, Mithapukhar and Badargani,
there have been little or no fissures or outpourings of sand and water, and no
contraction or filling up of river channels, tanks and wells. It is in the pali tract,
which comprises an area of about 3,000 square miles, that openingsin the earth,
outpourings of sand and water obstruction and filling up of river channels,
and filing up and breaking of pucka wells have taken place. The openings
are of two kinds, viz. (1) longitudinal fissures or cracks in the ground, extending
in many cases several hundreds of yards in length each, and gaping from
a few inches to 4 or 5 feet or more in width, and (2) circular holes, generally 4 to 5
feetin diameter, that burst open, throwing the surface earth several yards
away. Both kinds vomited sand and water. Openings of the latter descrip-
tion are more abundant in the Kurigram and Gaibanda subdivisions than else-
where. Thedirection of the longitudinal fissures was usually south and west
along the sides of tanks or rivers. In Saidpur and some neighbouring villages,
forming the south-west corner of the Nilphamari subdivision, and in some
scattered villages, such as Domer Bazar, Pochapukhar, etc., there were no fissures
or outpourings of sand or water. In some places in Nilphamari subdivision it
has happened that while one village was visited with these fissures and natural
fountains of sand and water, an adjoining village escaped them. In Kurigram
subdivision there is said to be notan acre of land without its fissures and out-
pourings of sand, more or less, of the description stated above.

[ "95 -)

320 OLDHAM: GREAT EARTHQUAKE OF 1897.

3. The fissures running parallel to rivers and tanks, along their sides, have
more rarely been accompanied by outpourings of sand and water. They appear
to be due to the weakness of the unsupported banks. Owing to them and other
causes, nearly all river channels in the pali tract have narrowed, some to the
extent of 10 to 20 per cent, of their width. In many instances this narrowing
process has crumpled up the bridges, or otherwise seriously damaged them.
Moreover, the rivers, tanks and wells in the pali tract more or less have been
filledjwith sand.Z Both large and small rivers suffered thus, and most of the
small drains have been (entirel y silted up. Small streams, 100 to 150 feet in width
have been filled up Şo to 75 per cent. of their depth. In large rivers the filling up
of channels or at least the main channel is not more than 15 to 20 per cent. of the
depth. The branch channels, however, in some cases have been considerably
filled up. The principal drainage channels of the district that have suffered most
injury are the two channels at Rangpur, the river Ghagat, into which they fall and
on the banks of which the Civil Station of Gaibanda stands, the Buri Tista nadi
passing by the important town of Ulipur, the Bamni nadi between Ulipur and
Kurigram, the Jerai nadi, the Kotakhali, the Manash, the Hathalia, the Hatia,
the Maroghagat, west of Gaibanda, the Naleya nadi, the Tista and the Mara
Akira. Several places inthese rivers which were not fordable before the earth-
quake, but could be crossed only by ferry, became fordable through the earth-
quake. The bed of the Ghagat referred to above, which passes through the
heart of the Gaibanda subdivisional headquarters, rose about Io to 15 feet from
Naldanga in the north to Badiakhali in the south, a distance of nearly 22 miles ;
the consequence of this has been that while the river is fordable,the ordinary
drainage has passed over the country, and a number of the villages on either
bank have been under water since the earthquake.

4. Of the 3,000 square miles of pali land in the district, the tracts that have
Suffered most are about Rangpur and in eastern and south-eastern parts of the
district, where about 20 per cent. of the area is said to be covered with sand. In
- the north-western and western tracts there have been fissures and outpourings
of sand and water nearly everywhere, but not on suchan extensive scale as in
the eastern and south-eastern parts. The outpsurings of sand have been greatest
in the jurisdictions of thanas Sunderganj, eastern part of Gaibanda, parts of
thanas Ulipur, Nageswari, Mahiganj and Katwali (Sadar).

Letter No. 1288 F., dated Dinajpur, the 20th October 1897, from N. Bonham- `
Carter, Esq. Magistrate of Dinajpur, to the Commissioner of the Rajshahi
Division.

In reply to your memorandum No. 706 Mcl., dated 23rd August last, for-
warding a copy of Government Order No. 4621 J., dated 14th idem, requiring
certain information on the subject of fissures inthe ground caused by the earth-
quake of 12th June, I have the honour to say that fissures of a minor nature
occurred; throughout the district. They occurred usually along the edges of
rivers or of low-lying land, and the largest that I have seen were not more than a
foot, or perhaps a foot and-a-half, wide. Some are reported to have. been a
mile long, but I have seen none as long asthat, though some are of considerable

{3e 9)

APPENDIX B 321

length. The fissures in most instances ran parallel with the bed of the river, as
if the whole bed of the river had sunk, but occasionally transverse fissures seem
to have occurred. Outpourings of sand and water also occurred fairly generally,
but in no great amount. These outpourings also occurred in the beds of rivers
or in low-lying land. I myself observed mud and water bubbling up near the
bed of the Purnabhaba river immediately after the earthquake. In this
instance, it appeared to me that there must have been, prior to the earthquake,
some water under the loose stratum of upper soil. The earthquake had the effect
of making this stratum of loose soil subside into the hollow occupied by the water.
The water was, consequently, forced up to the surface through the loose stratum of
soil. ‘he subsidence of the soil was indicated by the fissures in the ground
near where the cutpourings occurred. The police have supplied me with a list of
places where the fissures occurred, but I place no great reliance on their accuracy
or completeness. The list, moreover, conveys little or no information to any one
unacquainted with the localities named. I have, therefore, refrained fiom sending it.

Letter No. 554 G., dated Bogra, the 22nd Fuly 1897, from Umes Chundra Bata-
vyal, Esq., Collector of Bogra, to the Commissioner of the Rajshahi
Division.

3. In the Khtar or clayey portion of the district, no physical changes on the
surface of the ground due tothe earthquake are apparent, but in the
alluvial tracts the ground is cracked and riven in all directions. Long and
large fissures have appeared in many places, and sand and water oozed out of
them. The ground has been upheaved in some places, while it has sunk in others.
Beds of rivers and tanks have been visibly elevated in some places, while pools
have been formed on even ground in others. In the alluvial tracts the water of
wells thrown out and the wells themselves in many cases were choked up with
sand. Some reporters state that the smell of sulphur was felt coming out of the
fissures in the ground at the time of these seismic disturbances,

Letter No. 277 G., dated Bogva, the 13th September 1897, from Umes Chundra
Batavyal, Esq., Collector of Bogra, to the Commissioner of the Rajshahi
Division.

2. No river channels in this district have been completely filled up, but the
beds of some of them have been perceptibly raised here and there. A notable
instance of this took place in the bed of the Karatoya river, near the village of
Sultanganj, three miles south of Bogra. Here there was a deep pool in the
bed of the river, known as the Ramdaha, where in the driest season, the depth
of water (the villagers say) used to be about 15 feet, After the earthquake every-
body was surprised to find the river bed at the place to be almost dry. The
current of the river ceased to flow, and the villagers flocked to catch fish in the
mud. After a few days the river cut a small channel for itself over the place,
and there was no longer any obstruction to the current. In many other places
the people have remarked that the usual depth of water inthe Karatoya river
has decreased after the earthquake.

* (dap)

322 OLDHAM: GREAT EARTHQUAKE OF 1897.

3- In the course of my recent tours in thana Sariakandi, I was informed by
several intelligent observers that the flood water in this season used to submerge
the fields of village Ram Chandrapur before those of village Mathurapara.
This year, however, the case is reversed, and the people think that there has been
a relative upheaval or subsidence of the ground in one of the two villages.

4. Fissures in the earth and the outpourings of sand and water are confined
to the soft pali soil between the Karatoya and the Daskoba rivers. They are
absent in the hard khiar soil to the west of the former river. These fissures
were very numerous. The openings in the ground, however, seldom exceed two
feet in breadth, but their length varied very much in some cases, the same fissures
being observed over a mile or more. Their general direction was from north to
south. The width of the fissures increased in those places (very numerous also)
whence sand and water were ejected. These outpourings of sand and water are in
local language called bhorka. From my own house I noticed sand and water
oozing out of the ground on the opposite bank of the Karatoya, and a stream of
water was flowing down the sloping bank into the river. I, however, did not
witness the bhorkas (seen and described by many other people) in which the
water ejected rose several feet above the ground. Most remarkable is the story
told by one Rahaman Ali, Mondal of Bonepara, and 1 give the substance of it in
his own words—“ My house is on the bank of the Deakoba river. I was out to
see fields when the shock came. I could not stand erect, and I had to sit down.
The stream of the river was much agitated asif there was a strong gale. But
in a gale the waves move in one direction, on this occasion they moved in all
directions. The water seemed to swell, and for a time the surface was covered al!
over with foam, so that I saw only foam and no water I then observed a large
spout of water issue out of the river to the north-east. The column rose, I think,
so haths! over the surface, I see the mast of the Magistrate's boat (Note—It is 24
cubits in height). That column of water was much higher than this, and in cir-
cumference it was not less than 70 or 80 haths. This ] saw as the sound resembling
cannonade was heard. 1 ran towards my house in great terror from the char? As
I looked back, the column of water seemed to diminish. 1 did not look at it
again. A fissure has appeared on the west of my house, which 1 think is three
miles long and about two haths in breadth. Sand, charcoal, rotten vegetable
matter issued in large quantities out of the fissures.”

5. It is very likely that the witness exaggerated the height of the column of
@ater that he saw, and it is doubtful whether the said column of water
really rose out of the bed ofthe river or froma char in the river. But the
Sub- Manager of the Dighapatia State, Babu Rakhhyakar Moitra of Nowkhila, an
intelligent and educated man, and a member of the District Board, who happened
to be, at the time, in the char of the Deakoba river, saw springs of water rise to
the height of several feet above to the ground. He described them to be in ap-
pearance like small conical temples rising to the height of five or six haths. As
he ran towards the village, he was at one place surrourded on all sides with these
bhorkas. 1 have visited several places where these bhorkas appeared. Indeed

1 Ze, Cubits. ? Sandbank.
( 322 )

APPENDIX B, 323

on the other side of the Karatoya it is scarcely possible to go over a mile of
ground without meeting several such places. Small craters were formed out of
which sand and water issued with a peculiar sound, which the people described
as * hun hun."

Letter No. gooF., dated Pabna, the 20th September 1897, from C. A. Radice,
Esq., Offg. Magistrate of Pabna, to the Commissioner of the Rajshaht
Diviston. d

With reference to your memorandum No. 706 Mcl., dated 23rd August 1897,
forwarding for compliance the Under-Secretary to Government in the Judicial
Department, Circular No. 4621]., dated 14th August 1897, I have the honour to
report that, in the Sadar Sub-division of this District, the earthquake of-the 12th
June only caused the beds of a few of the dry river-beds, close to the big rivers,
to sink for a few inches. Cracks were consequently formed along the banks and
parallel to the bed. In the beds themselves, sand was, in places, projected to a
small extent. These disturbances, showing as they do, subsidence of surfaces,
below which the soil is merely lightly deposited sand, are all very slight.

2. In the Sirajganj Sub.division, the seismic effects have been far more
marked. I annex copy of the Sub-divisional Officer’s letter. This officer has failed
to be sufficiently precise, but in order to avoid further delay in the submission of
this report, I forward it in its incomplete state, with the suggestion that the
Officiating Director of Ihe Geological Survey of India be asked to refer direct to
that officer, if he desire to have further particulars. This procedure will save
time and labour and is, moreover, to be recommended, as the Subdivisional
Officer of Sirajganj is a well informed officer, well acquainted with his sub,
division.

Letter No. 3507., dated Stvajganj, the 17th September 1897, from T. Emerson,
Esq., Sub-divisional Officer of Sirajganj, to the Magistrate of Pabna.

In reference to your memorandum No, 791J., of 1st September 1897, I
have the honour to send the following notes of what I have seen myself of the
effects of which in'ormation is required by the Director of the Geological Survey.
The delay was occasioned by the non-compliance of the Kanungo and Sub. -
Overseer, Local Board, to report what they had seen.

1. Fissures in the town of Sirajganj were generally parallel to the nearest
water channel. In the Kutcherry maidan there were several fissures, from 12 to
18 inches wide. The ground invariably subsided on the side of the fissure
nearest the Katakhal channel. The subsidence was from 1 to 2 feet. The
greatest fissure about 23 to 3 feet wide and in the form of an arc of a circle was
in Messrs. Hannah and Co’s. compound. The roads were fissured in several
places, always at right angles to their direction. The road from Sirajganj to
Pabna was fissured in this way, and sankin many places. The road from
Sirajganj to Badraghat was split at various places and also subsided. The
chief fissures in this road were near the jute mill at the entrance to the town and
at Sealkole village.

At Alukdia and Narnia were fissures from 1 to 2 feet wide parallel to the river,

Y 2 poses

324 OLDHAM: GREAT EARTHQUAKE OF 1897.

On the Koijuri road, just beyond the ferry on the opposite side of the river -
from Koijuri, there was a series of wide fissures parallel to the river for about
200 yards. Thisis one of the worst series of fissures. The Belkuchi road is
fissured about two miles from Sirajganj; holes in parts 2 feet wide, further
south no injury.

Along the bank of the river near Nalka and Putas (the south bank of
the river) the earth was fissured for àbout two miles very badly, in a series of
parallel fissures extending to from 3 or 4 yards to 40 to 5o from the bank
of the river, the fissures varying from 1 to 2 feet width.

At Chanbaria, there were also fissures similar to those at Alukdia, parallel
to the river.

2. Outpouring of sand.—This has been greatest at Sirajganj. Sand was
emitted in great quantities in the Kutcherry maidan, by the side of the Pabna road
bevond Messrs. David and Co's house, in Messrs. Hannah and Co's. compound ;
also at Sealkole and Bhadraghat. I have not noticed it in other places, but have
been informed that there were رف‎ emissions of sand forming temporary
wells near Kamarkhonda.

3. Filling up of channels,—

(a) The Katakhal, just by the Barakuti, was considerably disturbed by the
earthquake, and sand poured into it from beneath.

(b) The Dhanbandi, the river which flows through the town, was being excavat-
ed at the time by the Steamer Company., Opposite Messrs. Hannah and Co.'s
Ghat two chars of emitted sand formed in the river.

(c) The Telkupi khal, which runs due west from Sirajganj, was half filled with
sand for about four miles. In an ordinary year it would not have been navigable
after such a filling up, but this year we have luckily had a record flood.

Letter No. 349F-—G., dated Calcutta, the roth October 1897, from G. Stevens
son, Esq. Offg. Commissioner of the Burdwan Division, to the Chief
Secretary to the Government of Bengal.

4. In the Raniganj and Sudder Sub-d'visions of the District, no phenomena
. of the kind, referred to in the extract from the Director of the Geological Survey

of India’s letter No. 693, dated 26th July 1897, are reported to have occurred.
* * * * * *

*

6. In the Katwa Subdivision some fissures are said to have been caused in
the bed of the Bhagirathi and the Ajai rivers by the earthquake. The greatest
width of a fissure was 2 feet, or a little above, and sand, water and dark coloured
mud were observed to issue from these openings to a height of 1 to 13 cubits

above water-level.
zm * * * * x ES

7(a). In the Kalna Subdivision a small fissure is said to have been formed in
the earth near a hill, in village Kobla, in thana Purbasthali, through which a small
quantity of sand and water issued. The spot being now under water, the fissure .
cannot be measured, but its extent was not great. No other fissure or outpour-

ing of sand and water was observed in any other part of the subdivision.
% * * * * * a

E 9

APPENDIX B. 325

o In thana Sakalipur water oozed from a fissure t5 feet long in a field near the
village of Tagbora. In the jurisdiction of the Lobpur Police Station there were
three cracks observed. One was at Balshanda, about 720 feet in length, and 3 to
9 inches in width at a distance of about840 feet from the river Majurakha ; the
second was at Tilpara about 840 feet in length and 3 to 3 inches in width, at a
distance of about 360 feet from the said river, and the 3rd at Gopalpur ahout
75 feet in length and 6 to 12 inches in width, at a distance of about 15 feet from
theriver Bakreswar. A small quantity of sand and water issued from these three
cracks for a short time. In the Moweswar thana one fissure occurred about 24
or 30 feet in length, south of the Kana river, in village Hatina, from which water

was seen to issue.
* * * e * * *

11. Four fissures in the earth are reported from the Subdivision of Vishnupur.
One was in the sandy bed of the Damudar at Rangamati, 11 miles north of the
town of Sonamukhi ; a second was in a ploughed field at Dhousimla, 5 miles east
of the said town, and a third was at Chowlia, 6 miles west of the said town on the
banks of the river Sali; while the fourth was in the bed of the Darkeswar river
about 6 miles north-east of the town of Vishnupur and 15 miles south of Sona-
mukhi. All the fissures threw up jets of sand and water with considerable force,
but it is difficult to give their exact size now. The jets were some 10to 15 feet
in height, it is said.

* * * * * * * *

15. There were several fissures in the earth in the villages of Bhagabanpur,
Bharatpur, and Nandeswar, all within the jurisdiction of the Debra Police Station
inthe Sadar Sub-division of Midnapur. Sand, water and mud were emitted
from these openings, but in no case was any damage done to the river channels or
tanks. The average length, breadth and depth of the fissures were 45 feet, 9
incres and 18 feet, respectively. Most of them have disappeared owing to the
rains.

d * * * * * * *

26. There were no fissure or landslips in the Hooghly District on the occasion

of this earthquake,
* * * * * * *

33. There were no fissures or landslips in the District of Howrah.

Letter. No. 2073F., dated. Bhagalpur, the rath September 1897, from H.F.
McIntosh, Esq, Magistrate of Bhagalpur, to the Commissioner of the
Bhagalpur Division.

With reference to your endorsement No. 103-P. W., dated the 25th August
1897, forwarding a copy of Government of Bengal’s No. 4620-J., dated the 14th
idem, regarding the earthquake of the 12th June last, I have the honour, in con-
tinuation of this office letter No. 1406-]., dated the 6th July 1897, to state as
follows.

2. The Subdivisional Officers of Banka and Madhipura report that no
fissure was opened in the earth in their subdivisions by the earthquake, nor have
any outpourings of sand or water or filling up of river channels or landslips been
observed anywhere,

(395

36 OLDHAM: GREAT EARTHQUAKE OF 1897.

3. In the Supaul subdivision it is reported that **a very small piece of land
on the bank of the river Kosi on the east of the village Guria, was cracked, and
only water issued out of it for one day." It is said that no damage was done
to any land.

4. In regard to the Sadar subdivision, I beg to submit a statement showing
the names of villages and the extent to which fissures inthe earth, outpourings of
sand and water, etc., were observed. .

Statement showing the information called fov in connection with earthquake
of the 12th June 1897—vide Commissioner's No. 1036, P. W.D.

Extent to which fissures to |Outpourings of sand

Name of village and police station. E و‎ cael

سا اا Pv i‏ ا

Gopalpur, Policestation Colgong, on | Length 4 or 5 rassis, breadth | Nii,

the bank of River Kalbulia. I cubit.
Villages Atania and Rassilpur, outpost | At places ; extent not given; | Nil.
Pirpainti, Police-station Colgong. willfollow on receipt of
report.
Village Immamnagger, outpost Pir- Ditto ditto Outpouring of sand
painti, police-station Colgong. and water.
Village Burgowri in police-station Ditto ditto Nil,

Sultanganj, bank of a pond in a guard
at the village opened.

Kalgawan in Police-station Sultanganj . | 3 or 4 bighas. Nil.

Letter No. 1444 G., dated Purnea, the 11th September 1897, from S. H.
Bernard, Esq., Collector of Purnea, to the Commissioner of the Gase
Division.

With reference to your memorandum No. 103-P. W. of the 26th ultimo,
forwarding copy of tbe Bengal Government, Judicial Department, letter
No. 4620-]. of the 14th idem, I have the honour to state that many fissures were
caused in this district by the earthquake. No more particular note was made of
their character than is given in paragraph 1 of the District Engineer’s letter
No. 180 of the 17th June last, a copy of which was submitted to you with this
office memorandum No. 780-G. of the 22nd idem. I observed several fissures.
They ran in different directions, and apparently more or less followed the course
of rivers, water-courses and roads. The configuration of the ground appeared to
r ule their course rather than the direction of the seismic wave.

3 ے6

APPENDI* B 327

Letter No. 761 G., dated Malda, the 27th September 1897, from F. H. Lea,
Esq., Magistrate ofMaidu, tn the Commissioner of the Bhagalpur Division.

With reference to your letter No. 103-P. W., dated 25th August 1897, for.
warding ccpy of Judicial Circular No. 4520-]. of the Government of Bengal, in
connection with the earthquake of 12th June, I have the honour to report that the
District Superintendent of Police and myself have made personal enquiries as to
cracks and fissures, etc., in the ground, the result of the earthquake, and that I
find that cracks opened out all over the low-lying or diara lands of the district,
varying from a few feet to half a mile inlength. The long cracks were generally
very narrow, from 1 to 4 inches, while some short cracks were several feet wide
and 5 or 6 feet deep. From these wide cracks in many places sand and water
spouted out, and from one fissure close to Gomastapur Thana on a char near the
River Mahananda, the Sub-Inspector of that thana states that he saw water and
sand and sulphur spouting out for 20 minutes after the earthquake was over.

2. | could not find that these cracks extended as a general rule in any
particular direction, but where they occurred on char lands, close to the Ganges,
Mahananda or Kalindri rivers, where they mostly did occur, they ran generally
parallel to the river. In a few places also highlands in the vicinity of the rivers
sank a small distance. No fissures or other surface disturbances were observed
in the highland tracts of the Darind.

Letter No. 3266 R., dated Dumka, the 28th October 1897, from R. Carstairs,
Esq., Deputy Commissioner, Sonthal Parganas, to the Commissioner of the
Bhagalpur Division.

2. In a village called Satar in Taluk Rohini, Subdivision Deoghur, there was
a fissure of about 18 cubits in length and 4 to 6 inches in width from east to west,
from which gushed a small quantity of sand and hot water. There is a small
stream passing by the place where the fissure took place, but no change in it has
been observed.

a In village Karharia, in the Godda Subdivision, there appears a fissure
in the earth. It was on both sides of a river and atright angleto it. Sand and
water outpoured from it. Before the earthquake there was no water in the bed of
the river, but after the fissure had tormed it was observed that water was trickling
down from the bank of the river. lt was also seen that the water was oozing out
from the ground. The fissure was 120 yards long and 4 to 5 inches wide. With
a stick of 3 cubitsits bottom could not be found. There were also little cracks
about the place.

4. In no other subdivision was any appearance of fissure reported.

5. T here were no landslips in the hills.

Letter No. 547G., dated Bankipore, the 20th September 1897, from S9. 4,
Bourdillon, Esq., Officiating Commissioner of the Patna Division, to the
Chief Secretary to the Government of Bengal.

4. In Bihar the public buildings developed cracks. The clock tower in the
Bayley serai was seriously damaged, and was subsequent!'y condemned by the

(2327 -)

323 OLDHAM: GREAT EARTHQUAKE OF 1897.

District Engineer. There was a fissure inthe cutcherry compound 50 feet long
by 1 foot, which is said to have absorbed the heavy rain which followed the earth-
quake, and had to be filled up. The hot springs at Rajgir are reported to have
discharged discoloured water for three days. |

* * * * * * *

11. Except in the district of Patna, as described above, no such phenomena
as fissures in the earth, outpouring of sand and water and filling up of river
channels were observed in any of the districts, nor has there been any landslip in
the hills.

Letter No. 101G.-Mis., dated Calcutta, the 22nd October 1897, from E. V. West-
macott, Esq., Commissioner of the Presidency Division, to the Chief Secretary
to the Government of Bengal.

With reference to your letter No. 4620J., dated the 14th August 1897, and in
continuation of my No 95G.-Mis., dated the 11th September following, on the
subject of earthquake of the 12th June 1897, 1 have the honour to report that no
fissure in the earth, or outpouring of sand or water, or filling up of river channels
were observed in the 24-Parganas or Khulna.

Meherpur Subdivi- 2. In Nadia fissures or cracks

A Subdivision, . :
adar Subarv sion. were observed in the places
In the neighbourhood

of Shikarpur and Hogul- noted in the margin, and sand‏ جو سے
KEE Bil. baria, along the course or water spouted forth from‏
them. No upheaval of river‏ ود Gurgaria River bed is the Metal‏

close to Muragacha. HO. Ie AE, tee cig acc >
and chars of that river. beds is reported to have occur-

red in any place in Nadia, Jessore or Murshidabad.

3. Two small holes, four feet in diameter and four feet deep, were observed in
village Dudsur, thana Sailkupa in the Jhenidah subdivision of the district of
Jessore. In the Magura subdivision cracks were observed on the banks of the
River Madhumati, near a place called char Bonni. Sinking of land is also
reported from the following villages :— ;

Name of villages. Depressions.
Sarsuna . . . ۰ ۰ . . g x6 x4
Satapara A : : - : ` - : 7l. x 7}! xo!
Parsarsuna . e . . . . e . 6 x6 X6
Gazirganti e . . e . e . . 18: X6 x 4M
Bagdanga > : . : . . " . 6 x 4M. x13!
Sablat . - - - e e . . e qi’ x 74 x3
Amean . ’ . e TD . . - 63’ x 43 x43

Sabekkhatore . > ^ : . ^ e ^ 2X 6 RG:
Satakhall . . . . . D e e 7 x 7 xa

4. In the Sadar subdivision of the district of Murshidabad, fissures accom-
panied by discharge of sand and water were observed in nine places in Thanas
Goas and Jellinghi, These were chiefly in marshy lands and in old river beds.
The fissures were generally of considerable size, the longest observed being ata
village named Bhuttomintolah in thana Jellinghi, and measuring two miles long

(«323 ) *

APPENDIX B. 329

and from two feet to half an inch wide. In the Kandi subdivision, about 25 `
fissures discharging black clay, sand and water to be observed near Chandipur
in thana Gokurna. These fissures are said to have been from 200 to 600 feet
long and from three feet to half an inch wide. In the Jangipur subdivision
fissures were seen in the bed of theriver Bhagirathi discharging black sand.
There were also cracks in a field near Rajanagar which discharged black sand.

Letter No. 3127, dated Dacca, the 15th November 1897, from the Magistrate
of Dacca, to the Commissioner of Dacca Division.

In continuation of my No. 2140, dated 20th July 1897, and with reference
to your No. 1300, dated 26th August 1897, I have the honour to forward a state-
ment showing thana by thana the information asked for by Mr. Oldham.

2. I beg to apologise for the delay which has occurred in submitting this
report, which was due to the papers being mislaid by my office.

FISSURES IN THE EARTH.

SABHAR.— Public road about 16 haths long, near village Kalinagar has

subsided.

Karasia.—Several dhan and jute khets in village Nandia have sub-
sided.

NAwaBGANJ.—Land from a place of the field in village Baldi to the Za
about 13 feet broad, has subsided.

RuPcaNJ.—Some places in villages Hargaon, Chaipara, Ballabdi,
Shukerail, Badburpur bil, Char Rahimuddi, Natubdi, Tarail, have
subsided.

NaRSINGDI OUTPOST.—Several places from Kharua river to Hatidoa
river and on the side of the Narsingdi road have subsided. Some

_ lands in village Char Bhagalia have subsided.

MANIKGANJ STATION.—Some parts of river bank in village Bhadar have
cracked. Some pieces of land in village Shaitya have subsided.
Some parts of land in village Samabari have subsided.

MuUNSHIGANJ,—Several places in villages Char Balaki, Char Kajuli and
Baisagram have been cracked. Several pieces of land in village
Raipur have subsided.

SRINAGAR.—A piece of lend, about 2,000 haths in length and 3 haths in
breadth in village Janginagar, has subsided.

A piece of land, about 40 haths in length and 40 haths in breadth
has subsided.

OuTPOURINGS OF SAND AND WATER.

KERANIGANJ.—In village Jinjira, land 6 haths long and 13 haths in breadth
has cracked, and black sand and water was thrown up. In some
seven places in a char of the Dhaleswari river, about 1 or 13 pakhi
ofland have subsided 13 to 3 feet; water and black sand were
thrown up. About-3 pakhis of land in village Washpur have

( 329 )

OLDHAM: GREAT EARTHQUAKE OF 1897.

subsided ; water, sand and clay were thrown up... About 3 pakhis of
land in Kamrangi char have subsided. Two or 3 pakhis of land in
a char in the Dhaleswari river near village Abdulpur have subsided
into the Dhaleswari river; land about 80 or 9o haths long and 4
haths broad has subsided ; water and sand were thrown up.

JAIDEBPUR.—Some lands in village Baraitali have been cracked; black
sand was thrown up.

SsBHAR.—Some portions of land on the south and western side of Kalia-
kari outpost have cracked; water and sand were thrown up; and
also in villages Nayanagar, Akura, and Hijaltali. i

KaPasrA.— Three houses in village Khirati have subsided ; water and sand
were thrown up ; same Occurrence took place in village Nandia.

Five houses in village Aral have subsided ; water and sand were
thrown up.

NAWABGANJ.— About a mile long and 1 foot broad of land in village
Gazikhali has cracked; large quantity of water and sand were
thrown up; and about 2 pakhts of land were covered with sand.

About 2 pakhis of land on the bank of the Dhaleswari river near
village Malikanda have subsided, and formed a deep tank. About
3 kanis of land in village Char Ambaria have subsided ; water and
sand were thrown up. Some damages were done to villages Nagor-
kandi, Baliapara, Narsinghpur, Nurapur and Mamdabad.

A khal about 50 haths in length has been formed in village Baidar-
bagh. A small Khal has been formed in village Kaitakhali.

RurcaNJj.—Some places in village Naopara have cracked; water aná
sand were thrown up A £Aal has been formed in village Gadai
Char. Some places in villages Baganagar, Haregaon, Darikandi,
Khadimpara, Jhaogara, Jagardi, Hargaon, Dutterbari have cracked,
sand and water were thown up; and about 30 haths of land in
village Kalagachia have been cracked with sand.

NARSINGDI OUTPOST.—Some lands in village Satirpara have cracked;
water and sand were thrown up; both sides of the Baher Char
khal have subsided ; black sand was thrown up.

Raipura.—Several places in the jurisdiction of Raipura Station have
cracked, and water and sand were thrown up. |

MowoHoRDIi.— Several places in villages Char Bagar, Kachikata, Dairer
Par, Tamakkandi, Monohardi, Baladi, Sukundi, Daibardi, Arjun
Char, Nama, Madhupur, Mirzapur, Kutubdi, Dudibil, Anundpur
Rampur, Lakpur, Narandi, Harirampur, Parpur, Chalak Char,
Kirtibashdi, Simalia, Char Doulatpur, Paratali, Rudradi, Hakibpur,
Maijdia,* Asadnagar, Gazaria, Chundunpur, Bengail, Sajardi and
Char Kasimnagar have cracked ; water and black sand were thrown
up.

Mk ees STATION.—À piece of land near Jaigir Bandar has cracked ;
water and sand were thrown up.

Several pieces of land in village Biswanathpur have cracked; water

339 )

(

339

APPENDIX B. 331

and mud were thrown up. Some damages occurred in village Ukra.
Some pieces of land in village Char Hijli have subsided; water
was thrown up.
Some pieces of land in village Takuni have cracked ; water and sand
were thrown up.

GHEOR ouTPosT.—Several pieces of land in the jurisdiction of Gheor
outpost have cracked ; water and sand were thrown up.

HARIRAMPUR.—Some pieces of land in Baragarabil have cracked; water
and sand were thrown up.

MUNSHIGANJ.—About one kani of land in village Raghur Char has subsid-
ed; water was thrown up.
A piece of land in village Alipura has cracked, and water was thrown
up. About one kani of land in village Bhater Char has subsided ;
water was thrown up.

Letter No. 1330%., dated Mymensingh, the and November 1897. from E. B. Harris,
Esq., Magistrate of Mymensingh, to the Commissioner of the Dacca
Division,

With reference to your No. 1300-G., dated the 26th August last, forwarding
copy of Government letter No. 4620, dated the 14th idem, I have the honour to
refer you to this office No. 942-J., dated the 3rd July 1897, in which I gave an -
account of the damage caused by the earthquake of the 12th June 1897.

2. In the Jamalpur, Netrakona and Sadar subdivisions fissures were in.
numerable. There were a good number in parts of Kishorganj ; and none at all in
the Tangail subdivision. All the smaller rivers in the first three subdivisions had
their beds upheaved in places, particularly the Katcha Matia, Kharia and Mirgi
rivers. Most of the tanks and wells in these parts were more or less silted up.

3. As regard landslips, several have occurred since the 12th June 1897, the
majority of which no doubt have been caused by seismic shecks. Some, of course,
are due to constant rainfall. As no observations were made at the time, it is
impossible to distinguish landslips during the earthquake and those caused by
the subsequent rain. The Garo Hills, which border the district to the north, bear
marks of numerous landslips.

Extract from Notes on the earthquake of the 12th Tune 1897, by the Munsif of
Isvarganyz, in the District of Mymensingh, forwarded with his letter, dated
Isvarganj, the 20th August 1897, addressed to the Officiating Director, Geolo-
gical Survey of India.

Lë * * * * * *
(3) Formation of the springs of sand and mud.
During the earthquake these springs were formed at several places here.

I have seen with my own eyes springs forming upon the bank of the river.

The pressure with which the sand, mud and water were thrown up by these

springs was very great. At several places these were thrown up over ro feet

high. The sand and water were forced up at a very rapid rate, and the forcing

(ger )

332 OLDHAM: GREAT EARTHQUAKE OF 1897.

up continued even 24 hours after the earthquake was over. In consequence
there was something like inundation at several places. The fields were flooded
over and the water reached the houses of several people. The people were
afraid of the entire subsidence of the place. I have heard of the people of Gauri-
pur, Kalipur, Bhowanipur within this Munsifi, actually preparing floats to save
themselves from drowning. Besides many of the people were flying away in wild
confusion in all direction soon after the shock for the very fear of life. This
country is high; the Brahmaputra was long ago silted up; the people here had
thus no experience of inundation ; and hence that awful panic.

Another effect was that arable lands were eatirely covered over by the layer
of sand forced up through these springs. So much sand and mud were thrown
up that it is estimated that about 2 annas? of the land within this Munsifi, Isvar-
ganj has been entirely covered over with sand at places 3 to 4 feet deep, by
which these lands have become utterly useless for cultivation.

With reference to these sand springs it is necessary to note here that I had
occasion to notice the nature of thesubsoil here during the course of the excava-
tion of certain wells. I found that near the surface of the land there is a layer
about و‎ or 10 feet deep which is generally composed of loose sands. I found
both red and white sands in their composition. Below this there is another layer
of thick mud with sprinkling of sand. This layer is from 4 to 6 feet deep.
Below this there is a layer of semiliquid putrid mud. This layer is very thick.
A bamboo pole was forced into this layer. It went down some 15 to 20 feet
without much resistance. I believed this layer to be over 50 feet deep. It does not
appear what is below this layer. This appears to be the state of the soil through-
out this district. Throughout this district the fissures were formed only on the
upper sandy layer, and were in no place more than g or 10 feet deep. This also
explains why everywhere here mud and sand were thrown up in abundance.

Lelter No. 1074 G., dated Chittagong, the 13th October 1897, from F.R.S. Collier,
Esq., Officiating Commissioner of the Chittagong Division, to the Chief Secre-
tary to the Government of Bengal.

With reference to Government letter No. 4620J., dated 14th August 1897,
calling for certain information in connection with the earthquake of 12th June last, .
I have the honour to submit copies of the reports submitted by the Subdivisional
Officer of Brahmanbaria and the District Engineer of Tipperah, and to state that
no other fissures are reported from any other place in this division. No land-

slips have occurred in any district.
Report of the Subdivisional Officer of the Brahmanbaria,
THANA BRAHMANBARIA.

1. A slight fissure about 200 feet in length, 2 inches wide, and about 2 cubits

or more deep in Syampur char land, opposite to Akhaura, in thana Kasva,
THANA NASIRNAGAR. _
2. A khal about a mile long has opened at Chatalpur.
1 7, e,, one-eighth.

C7332")

APPENDIX B. 333

Tuana Kasva.

3. Three fissures about 500 cubits long about 2 inches wide and about half a
cubit deep in village Kharampur. They poured out sand. Many small ditches
appeared and poured out sand. Long, wide and deep fissures were seen
in the surrounding char lands near Mogra. Sand and water were spouted out.
Some portions of land subsided and at points the subsidence was about 5 inches
below the normal ground level.

THaNA NABINAGAR.
4. A fissure about 300 cubits in length, about 4 cubits wide, and about 23
cubits deep at Madhupur. It threw up Sand.
2. Nolandslips were noticed in this subdivison.

Report of tha District. Engineer of Tipperah.

Begs to state that there has neither been any fissure in the earth nor any
la1ds!ip within the area of this district, except the tapping of a big spring about
8 inches in diameter in a pucka well under construction at Chandpur, emitting
sand and water and filling in the well. :

(a) A long fissure about 500 feet in length and average 18 inches wide on
the road leading to Agartala, about one mile from the British
boundary. Direction of crack, east to west.

(b) Several fissures in the earth at Agartala of the above description and
direction, and out pouring of sand only in some place, only to the
height of the level of the ground.

(c) One thatched shed sank down half of its length about 30 inches.

(d) The water of a tank was 6 feet deep before earthquake, but only 3 feet
after it,

Letter No. 1592XX.—8G., dated Dhubri, the 8th September 1897, from A. G.
Hallifax, Esa., 1.C.S., Deputy Commissioner of Goalpara, to the Secretary
to the Chief Commissioner of Assam.

With reference to your memorandum No. 5148-59G., dated the 9th August
1897, requesting to supply certain information as to extent to which fissures in
the earth, outpouring of sand and water, etc., have been observed on the occasion
of the last earthquake, I have the honour to state that fissures in the earth were
observed throughout the Dhubri and Goalpara towns, more especially on the
river banks. There is not a single road in the town which escaped damages
caused by fissures inthe earth. The earth was crackec in all directions. The
outpourings of sand and water were also observed in a good number of cracks, .
both in the towns and in the interior of the district. The effects of the earthquake
in the interior of the subdivision were similar to those experienced in the town.
Most of the wells have been filled up, and the ground in various places has been
covered with sand and water. Many of the roads in the interior have been
damaged owing to cracks. Reports have been received that a good number of `

(683847)

334 OLDHAM: GREAT EARTHQUAKE OF 1897.

river channels has been filled up with sand, whereby some ferries have been tem-
porarily closed. Landslips were found at three places on the Bhumesvar Hill, but
caused no damage to the crops, as the land below it is covered with grass jungle.

Letter No. 2675, dated Gauhatt, the 26th November 1897, from Captain P. R. T.
Gurdon, I SC., Deputy Commissioner of Kamrup, to the Secretary to the
Chief Commissioner af Assam (through the Commissioner, Assam Valley
Districts).

With reference to your memorandum No. 5148-59G., dated the oth August
last, asking for a special report on the extent of fissures, outpourings of sand and
water, the filling up of river beds, etc., owing to the earthquake, I have the honour
to submit the following report regarding the Gauhati subdivision ; a separate
report by the Subdivisional Officer of Barpeta is submitted with its enclosures.

2. You are aware that in several places in this district deep fissures were
formed owing to the earthquake. In the Bordwar Tea Estate, a garden situated
at the foot of the Khási Hills, two fissures were measured by Mr. Gilman. They
were found to be 40 and 45 feet deep, respectively. The fissures varied in width
from 2 to 4 feet. It is observable that the fissures that opened near rivers were
in a direction parallel to the banks. The strand road from Sukleswar ghát to
Bhorolumukh furnished a striking illustration of this.

Throughout the length and breadth of the sadar subdivision cracks were very
numerous. The road from North Gauhati ro Tambulpur, notably the portion
between Kamalpur and Rangia and that between the 5th and 6th mile of the
Rangia-Tambulpur road, and the road to Hajo from Tambulpur vif Barama and
Nalbari, and the adjoining fields, were fissured in several places. In many
instances cattle are said to have lost their lives by falling into the cracks.

3. As regards effusions of sand and water, it is reported that in nearly all the
tahsils the rice-fields opened in fissures, and through these were ejected sand
and water, which sometimes rose to a height of several feet, and on falling formed
sand craters. I may mention that I personally saw these sand geysers in the
rice-field of Nalbari, where I was in camp at the time of the earthquake. The
geysers rose to a height of from 3 to 4 feet. The water was not hot, nor was there
any sulphurous smell, as has been reported by some correspondents. I should.
like to draw attention to the fact that earth-waves were distinctly visible from the
resthouse compound, which is situated close to a large pathar or series of rice-
fields. The waves could be seen following one another at intervals, the ghz rice
falling and rising as the waves progressed. The sight was a most extraordinary
and interesting one. I have forgotten to state that the sand geysers were subse-
quent to the great shock and continued for quite half an hour. In some places
roads were rendered impassable, and the houses became uninhabitable on account
of the sand that was forced up.

4. River beds were raised and rendered shallow by sand emitted from under-
ground and by the upheaval of the beds. Several rivers, formerly very deep, have
now become shallow and fordable in places. The Ghoga stream, on the Tambul-
pur road, which was formerly very deep, had its bed silted up, and could easily
be forded. The beds of the Pagladia, Dimila, Motonga, Borolia, Chenguri, the

( 334 )

APPENDIX B. l 335

Hajosota, the Barnadi, and many hill streams in Tambulpur have been consider.
ably raised, and the water from them overflowing the banks inundated the whole
country on both sides. Interior trafic and communication by river was retarded
owing to the shallowness of many of the rivers. The beds of irrigation canals in
Tambulpur were raised and the water stopped.

5. Landslips are reported to have occurred in this subdivision. Near Chouki,
5 miles to the north of North Gauhati, a small landslip occurred from a hill on
the roadside, but not so as to render the road impassable. Further to the north,
near Chutiapara, huge rocks feil down the Kohora and Deoduar Hills. In Hajo
several rocks have rolled down the hills, and the temple of Siddheswari, which
stood on one of these hills, is no longer in existence.

6. In. Bhutan the shock of the earthquake seems to have been certainly
severe, if we can judge from reports. A portion of the Gomchichitong hill to the
north of Dewangiri fell. The Bhutan boundary pillar at the foot of this hill
was broken, and the high land near it subsided. A portion of the road from
Dewangiri to Oranggaon, within the jurisdiction of Bhutan, was blocked up by a
landslip from the Orang hill. People having to pass by that road now takea
circuitous route by the bed of the Dia stream. The Ridaja Hill to the north of the
Orang hill is reported to have been “levelled to the ground.” This hill was
formerly verv high, its summit being covered with snow nearly all the year round.
The report is, however, hardly credible.

The road leading to Susa town was blocked up near Solikhar by the fall of
Nadangsing and Jirim hills. Foot passengers can still pass by the road, though
riding is impossible.

Portions of the Oangla and neighbouring hills, with the houses of the Lama,
fell down. The iron bridge over the Monas is reported to have collapsed. The
principal mantraghar (or place of worship) of the Desa Raja and the stone house
of Gorajung Raja were levelled tothe groui d. Besides these, there were numer-
ous landslips within the jurisdiction of Bhutan, among smaller hills, which have
blocked up roads and rivers in places.

Letter No. 366, dated Barpeta, the 23rd September 1897, from Babu Madhub
Chandra Bardalai, Subdivisional Officer of Barpeta, to the Deputy Commis-
stoner of Kamrup.

With reference to your memorandum No. 1613, dated the 17th ultimo, forward-
ing copy of the letter No. 6(2, dated Calcutta, the 26th July 1897, from
R. D. Oldham, Esq., Officiating Director, Geological Survey of India, to the
Secretary to the Chief Commissioner of Assam, I have the honour to report as
follows.

2. That immediately after the occurrence of the earthquake of the 12th June
1897, the earth opened in fissures in every direction, and as far as it has since been
ascertained by local enquiry, everywhere within this subdivision, the openings
being wider and greater in length in places higher up than in places lower down
near to the Brahmaputra ; while on the other hand geysers rose, throwing up sand
and water from different heights ranging from 2 feet to 12 feet. Bits of coal,

C 35)

6 OLDHAM: GREAT EARTHQUAKE OF 1897.

old logs, and other vegetable substance came out of the springs with the sand and
water. The springs were more numerous in low lying places than in high tracts
towards the north near the Bhutan hills.

3. Simultaneously, wi h the springs, the river beds were raised high as to make
them nearly of the same level with the banks, with the result that the volumes of
water imbedded hitherto in the rivers, added to the subterranean water thrown out
of the springs, overflowed all the places; high and low, the flood rising to the
highest recorded flood level the day after the earthquake.

4. That from the undulations observed, specially in the roads, it is pretty
certain that the wave of the force of the earthquake moved in the direction of
south-west to north-east.

5. From the subsequent high floods that have inundated the subdivision, there
could not be much doubt that the river channels have been raised, and that the
lands near the river banks have generally subsided to a depth varying from 3 to
7 feet or more, while the tracts of the subdivision towards the north nearer the
Bhutan hills remain higher as before.

I beg to enclose herewith, in original, a detailed report submitted by my Sub-
Deputy Collector, Babu Abhya Sanker Guha, who was specially deputed to make
local enquiries and report on the points in question, for your perusal.

Memorandum, dated Barpeta, the 19th September 1897, by Babu Abhya Sanker
Guha, Sub-Deputy Collector.

The following report is submitted to the Subdivisional Officer, Barpeta, with
reference to memorandum No. 1613, dated the 17th August 1897, from the
Deputy Commissioner of Kamrup.

When the earthquake of 12th June occurred 1 was out on tour in the Paka
Mauza. The very night of the disturbance I went by boat from Kahikuchi to
Lachima. Since then I have travelled by land and water in the rural tracts all
over this subdivision for the purpose of enquiring into the condition of the people
and crops and damage done by earthquake. I have thus been able to observe
personally the effect of the great shock over wide areas in the shape of fissures in
the earth, outpouring of sand and water, and filling up of river channels, etc.
1 have also made enquiries on the subject, from which 1 learn that the phenomena
above mentioned have been widespread so far as the Barpeta subdivision is
concerned.

z. The fissures have been of different kinds in different localities. In some
places they have been many and extensive, tearing up the ground in a very rough
manner ; in others they have been few, and their size small. In several places
rice-fields and public roads have been badly injured by them, some roads have been
rent lengthwise, while on some deep gaps have occurred crosswise, letting in strong
currents of water from the submerged fields. In several places again the earth
has subsided more or less alongside the rents, and this has been specially marked
on the banks of rivers and streams, such as the Kaldia and Singra, the Chaul-
khoa and the M anás.

(336. )

APPENDIX B. 337

3. Immediately after the earthquake we took refuge on an open field not far
away from a Nepali basti on the Singra river, and there the earth opened in wide
fissures, emitting sand and water. This took place so quickly and in so many
directions that we had to make a precipitous retreat into our boats. This sub-
division having been extensively flooded immediately after the earthquake, the
fissures cannot be observed so well as in the higher districts; but that they have
been widespread is evident not only from what can be seen in those tracts which
are dry, but also from soundings taken in water. In travelling by country boat
my boatmen have found varying depths at close intervals while pushing the boat
by means of bamboo poles. Also in wading through water on foot (I was obliged
to do this, as boat journey was not everywhere practicable), I frequently slipped
into deeper gaps, thereby showing that the soil below the water was fissured more -
or less, Inthe northern and elevated parts of this subdivision the cracks are
still seen, although they have been filled up to some extent by earth washed away
by rain or obliterated by cattle. I enclose herewith a rough sketch (not printed)
showing some of the fissures in villages Nuntola and Bairapur in the Roha tahsil
and a statement showing those in the villages Bhaluki, Balipara, Pakaketkibari,
Tuplipanbari, Nityananda Panbari, Raipur, Kukuabatabari, Dalaigaon, Muguri,
Haguri, Sariha, Chakla and Patacharkuchi in the Bajali tahsil. The length,
breadth, depth, and directions of the fissures are mentioned.

4. As already stated, I was in a boat at the time of the earthquake, and the
country being open on all sides, I could clearly observe the springs pouring
out sand and water. l saw several spouts on the Singra River itself, one of
which rose about 7 or 8 feet high, close to our boats. The boats escaped
providentially. Some of the crew and the supervisor kanungo of the Barpeta
circle, who had accompanied me, jumped overboard in great terror, and
managed to reach the shore. Some geysers rose close to the beach, and
several were seen further inland, like fountains playing. From their distance
I think some were as highas ro to 12 feet. Theroar produced by these some-
what resembled that of. a bore. These outpourings continued till a couple of
hours after the earthquake, but their force gradually faded away as the country
was flooded. However, nearly two and half months after the great earthquake
I found water oozing out, though in minute quantities, from some of the sandy
craters caused by the geysers of 12th June in Mauza Domkachakabansi. A villager
told me that water came out of these whenever there was a severe shock. From
enquiries I have learnt that geysers have been common all over the subdivision.
As the water-bearing strata were forced up, the result was the immediate flooding
of the whole subdivision, high and low portions alike. The water soon subsided,
however, in the higher parts, and as it flowed down to the lower tracts, the in-
undation in the latter increased, which has been further augmented by two successive
floods, the third higher thanthe second, Bits of coal, old logs, and other veget-
able substance were forced up with the springs, exciting the wonder of the ig-
norant villagers. Rice-fields, wells, tanks, mullas, and creeks were covered with
sand ; and houses were damaged or destroyed where the geysers occurred inside
or close to them. Granaries containing paddy have also been injured or de-
stroyed in this way.

e | (42277)

338 OLDHAM: GREAT EARTHQUAKE OF 1897.

5. One of the most curious phenomena resulting from the earthquake of
12th June was the lowering of high, and the raising of low places. Many localities
which never used to be under water before were flooded immediately after the
earthquake ; on the other hand, beds of rivers and Als have been raised. This
raising has been affected in three ways; one by sand deposited by geysers
another by subsidence of the banks, and the third by upheaval. That there has
been actual upheaval of river beds, etc., can be inferred from the fact that portions
of several roads have been raised in a curious fashion. Undulations are seen on
the surface of roads which are not under water, and several bridges have been
raised owing to the timber posts which supported them having been forced up
by the upheaval. One such bridge is to be found on the road leading from Tenga-
gaon to Roha, where the centre somewhat resembles the apex of a triangle.

6. The Singra was a deep stream before the earthquake, and we could not
generally fathom it with the bamboo poles, about 20 feet long, which we carried
for pushing the boats. But after the earthquake it became so shallow that our
canoes, which drew about a foot of water, grounded in several places. In the
Kapla bil, in the Sarukhetri mauza, my kolong boat, which drew about 2 feet of
water, grounded several times. I learnt that the Kapla was a very deep bil before
the ہت‎ ete I observed the same phenomenon in the Chaulkhoa, the Mara
Manas, and the Kaldia rivers. In the Bajali tahsil 1 found two or three tanks
quite dry, the interior having been filled np by silt and raised to a level with the
surrounding banks.

Letter No. 939-G., dated Tezpuv, the 26th August 1897, from Lieutenant-Colonel
M.A. Grey, B.A.,1.C.S., Deputy Commissioner, Darang, to the Secretary to
the Chief Commissioner of Assam (through the Commissioner, Assam Valley
Districts).

4. In the Sadr subdivision there were fissures everywhere in low-lying
lands, and in several places the cracks were several feet wide and deep on the
west side of the Bharalu river. Large quantities of water and sand were ejected
from the ground. The Tezpur-Balipara tramway line was very Severely injured
in this way between the Sessa and Rangapara Stations. I visited this portion
of the district on the two following days—the 13th and 14th June—and can only
say that the extraordinary way in which the rails were thrown about, and twisted
into all manner of curious shapes, almost baffles description. The whole of the
low lying tract of the country about the Rangapara station wasa mass of little
geysers, out of which sand and water had poured out. In some places the jets
of water were reported to have been from three to four feetin height. The sand
thrown up had a peculiar smell, like sulphur, and in some parts was of a black-
ish colour, though as far as my personal experience has enabled me to speak, the
ejected sand was all white. The same thing also occurred pretty well all over
the Mangaldai subdivision. In many places the deposits of sand covered
several bighas of land. The whole of the Chapori near Mangaldai itself was
completely riddled with fissures, some being more than 200 yards long. The `

GE M

APPENDIX B. 339

deposits of sand were much greater on the Chapori than elsewhere, and covered
the land in some places to a depth of four feet. At Singrimari tea garden a foul
smell came from the sand and water spouted up, as well as sand. The farthest `
case eastwards of the earth opening and sand and water being poured violently
up, was at the Singlijan tea garden, near Helem, about 50 miles east-north-east
of Tezpur. There the water burst out actually in the midst of the coolie lines, and
was thrown up several feet high. This is remarkable, as being an isolated case
so far east of Tezpur. The garden in question is situated at the very foot of the
Duffa Hills.

s. The beds of some of the rivers were also thrown up several feet by the
earthquake. A smallriver at the Tangani tea garden ran dry and filled up slowly
during the night. The bed of the Daiphang River southward remained dry for
five days after the earthquake. The Nonai rose suddenly and flooded some villages
near Nalbari. The Kulsi flooded portions of mauzas Barantola, Dalagaon, in-
cluding Bholabari tea garden, and submerged the main road from Kalaigaon to
Paneri. The Diggaj did the same.

Letter No. 720, dated the 7th Fuly 1897, from the Assistant Engineer, Mangaldat
Subdivision, to the Executive Engineer, Central Assam Division.

As it may be of some scientific interest to ascertain the depth from which the
sand and water were forced up the fissures in the ground during the great earth-
quake of the atb June last, I have the honour to report, in continuation of this
office No. 690, dated 30th June 1897, that at the moment the earthquake was felt,
I was supervising the sinking of Dalgaon well. It was between 5-20 to 5-25
P.M. according to my watch, The well was sunk to a depth of 25 feet below ground
just before the earthquake, and after the earthquake it sank 6 inches of itself giving
a total depth of 253 feet. The sand met with at depth 25 feet was coarse river
sand with pebbles. The next day when I sent down the dredger, it stuck into sand
and it was with great difficulty that it could be extracted, when very fine blue
sand with a slight mixture of clay was brought in the dredger. Thissand had
filled in the wellto a height of nearly 4 feet, on removing the sand up to the
bottom of the well curb and a foot below it, the same coarse sand which I was
working before the earthquake was found and the well sinking was stopped at
sanctioned depth of 27 feet below ground. The depth of fine sand which has
been brought up the fissures in the ground appears therefore to be not less than
27 feet.

2. The pucka well in the Subdivisional Officer’s compound, Mangaldai, which
was sunk to a depth of 33 feet below ground had been filled in with the same
fine sand to a height of 16 feet above the bottom of well curb and the brick
steining cracked badly. This sand I am now removing for examination of the
well, which has suffered badly, inasmuch as the lower portion of the steining is
holding an inclined position and a portion just below the top has bulged out.
The depth of fine sand according to this well is not less than 33 feet below ground

as | ( 389 )

340 OLDHAM: GREAT EARTHQUAKE OF 1897.

Letter No. 686, dated North Lakhimpur, the goth October 1897, from H. A. C.
Colquhoun, Esq., B.A., I.C.S., Subdivisional Officer, North Lakhimpur, to the
Deputy Commissioner of Lakhimpur.

With reference to your memorandum No. 492G., dated the 21st August 1897,
I have the honour to forward a report supplying the information available.

1. (a) Fissures in the earth and outpouring of sand and water were very
general throughout the subdivision west of the Subansiri. They occurred gener-:
ally on the banks of rivers and on low lying or swampy ground. The fissures
were usually of a few inches in breadth and two feet in depth, and often exceeded
100 yards in length. The emission of sand and water was accompanied in
most cases by sulphurous smells. General subsidence also occurred in parts
affected by fissures. This is especially noticeable along the left bank of the
Dikrang for four miles above Narainpur. The bank has been deeply fissured, and
the surrounding country inundated owing to the subsidence of the soil. At
Kadam tea garden, again, an acre or two of tea has been destroyed owing to the
soil having become water-logged, and the garden has been subjected to unusual
floods. It is probable that these floods have been caused infact by silting up
of the Dikrang and Kadam river beds, but I am not prepared to say to what ex-
tent. That considerable subsidence has occurred is, however, unquestionable.

(b) Neither fissures nor subsidence have been reported from the three mauzas
east of the Subansiri.

Letter No. 749G., dated Sibsagar, the 8th September 1897, from Bernard V. -
Nicholl, Esq., I.C.S., Deputy Commissioner of Sibsagar, to the Director of
the Geological Survey of India, Calcutta. `

The following is a summary of the information received from the tahsildars
subordinate to me.

The occurrence of cracks in the earth from which sand was thrown up, similar
to those reported from Jorhat and Golaghat, also took place in the western
portion of this subdivision.

The spots where these phenomena were observed are all in the neighbourhood
of water courses. Some people were driven from their houses by the appearance
of sand and water coming up through the floor. A considerable portion of one
road in the same locality sunk, and will have to be thrown up again to raise it
above water level. A crack described as over 200 feet in length and ro or 12 |
cubits in depth and running north and south made its appearance in another road.
A tank 1o feet in depth is reported to have been completely filled up with sand ;
while a low lying cattle-track has been converted into a raised path by the sand
forced up from below.

Letter No. 1125G., dated Forhat, the 1st September 1897, from B. C. Allen, Esq.,
LC.S., Assistant Commissioner, Jorhat, to the Deputy Commissioner of
Sibsagar.

] have the honour to submit a report on the points mentioned in paragraphs 1

( 340 )

APPENDIX B. 341

and 2 of letter No. 692, dated the 26th July 1897, from the Officiating Director
Geological Survey of India, copy of which was forwarded with your memorandum
No. 670-71, dated the 19th August 1897.

2. Fissuves in the earth and outpourings of water and sand.—This pheno-
nienon was not uncommon in this subdivision, but no record was made either
of the size or number of the fissures that appeared. In Jorhat town, inthe Dacca
patti, the earth cracked ın several places, and in spite of the cracks not being more
than 6 or 7 feet in length and 5 or 6 inches broad, water and sand were thrown
up inconsiderable quantities. In the Domgaon,a mile north of the town, the
soil was much broken up, and between one and two hundred fissures, varying
from a few inches to 15 or 20 feet in length, appeared. The largest of these
must have been nearly a foot broad. Here also there was a considerable effusion
of sand, water, and mud.

The largest cracks reported were those on the Ladoigarh road, near Nokachar
tea estate, one of which was nearly 2 feet in breadth, the road being an unbanked
one ; these cracks were probably due to a fault in thesoil. :

3. Noinformation has been received of landslips in the hills.

Letter No. 955G., dated Golaghat, the 9th September 1897, from Lieutenant
W..M. Kennedy, 1.S.C., Assistant Commissioner, Golaghat,to the Deputy
Commissioner of Sibsagar.

In reply to your memorandum No. 670-71G., dated the roth ultimo, for-
warding a copy of letter No. 692, dated the 26th July last, from the Officiating
Director, Geological Survey of India, directing me to furnish the information re-
garding the recent earthquake, called for in paragraphs1 and 3 of that letter, I
have the honour to submit the following reports.

Rangamati mauza.—A fissure, 3 miles long and from 1 to3 feet broad, was
formed from the old Garhali to the new road near the Dhansiri river. It extended
for about 3 miles, and sand and water were emitted from it to a height of from 1
to 2 feet. Near the Dhansiri river there was a fissure 30 or 40 nals in length and
6 or 7 zalsin breadth. The surrounding land has subsided 3 or 4 cubits. In
Mohora Bilota village near Gela Nadi,a fissure, 150 za/s inlength and 1 or 2
inches in breadth, was formed. In Garaimari chaparí a fissure, 200 nals long
and 5 or 6 inches broad, and on the west of the chapart, a fissure, 120 mals long
and 2 or 3 inches broad, were formed, from both of which sand and water were
discharged. Also at Baukula village a fissure, 250 za/s long and 4 or 5 inches
broad, was formed, from which sand and water were emitted. Ina chapari near
Tinsukia a fissure, 180 nals long and 4 or 5 inches broad, appeared, vomitting
sand and water. Ona chapari near Dirai suti the land was fissured, the length
of the opening being 210 nals, and the breadth 3 or 4 inches. From this also
sand and water were discharged, and also from a fissure 210 alsin length and 3
or 4 inches in breadth which appeared in Botokoi village near Bena bhil. In
this mauza there may have been other fissures too in other chaparis or in the
jungle which were not ncticed,

: ( 341 )

342 OLDHAM: GREAT EARTHQUAKE OF 1897,

Ahatguri mausa.—In Rangamuri field a fissure, 60 or 70 nals long, Y cubit
broad, and 8 or 10 cubits deep, was formed, from which broken pieces of wood
were discharged. Sand and water poured out from a fissure in Ramdhani bhil,
the length of the fissure being 40 or 50 cubits, breadth 13 cubits, and depth 7 or~
8 cubits. In Charani Alli field a depression like a pond was formed, covering
about 3 bighas of land, the depth of the water being 12 or 13 cubits. Broken
pieces of wood were observed floating on the surface; near the Brahmaputra and
the Lohit rivers several fissures were formed, and in some places the land sunk 8
or ہ×‎ cubits.

Namdoyang mauza.—In the Lákhirá] Khát of Karuabahi sastva a fissure
half a mile long and about 1 foot broad was formed, from which sand, water,
and broken pieces of wood were discharged, several bhils were filled up with sand
and other bhils were deepened. Mari, Dhansiri, and Dijahi and other fans were
in parts filled with sand.

In Bez and Bilotia villages some fissures appeared in the Sond

Athgaon tahsil.—Only one small fissure 2 mals long and half a cubit broad
was formed ; a small quantity of sand and water was emitted.

Golaghat tahsil—In Mauza Monkhoa, lot No. 5, one fissure was formed, in
length 100 mals and in width 1 or 2 cubits, from which water and small pieces of
wood are said to have been discharged to a height of 5 or 6 cubits, and in some
places the land sunk.

In lot No. 6 fissures are reported to have appeared at Salmara, Mohkuti»
Naparnua, Bholuguri, Butuli Khoa, and Hauntli Khend. ‘The dimensions of these
fissures could not be ascertained by the tahsildar, as the land is now under water,
but some of these are said to have been 24 or 25 zals long and 2 to 4 cubits wide.
Minor fissures are said tohave beenformed onthe banks of Nobhil in Khumlai
mauza, but to have been immediately filled up with sand. In this tahsil only the
lands bordering on rivers and bhils were fissured, the soil being sandy and no
fissures appeared in the interior.

Missamara mauza.—ln Borgohain chapart, in the jungles, some very small
fissures are said to have been formed, from which water and sand were discharged.
These alleged fissures are now under water owing to the floods.

As regards paragraph 3 of the Officiating Director's letter quoted above, I
am informed that no hillmen from the Himalayan tracts visit this subdivision.
In any ease they would not visit the plains till thecold weather sets in.

Letter No. 1000G., dated Kohima, the 14th October 1897, from Captain H. W.G.
Cole, I.S.C., Officiating Deputy Commissioner of the Naga Hills, to the Secre=
tary to the Chief Commissioner of Assam.

3. In the plains portion of the district, observations were confined to the cart-
road connecting Golaghat and Kohima which runs through the Nambor forest.
Here no measurements of fissures were taken, but they were said to be very
general, and the openings were filled up by road gangs as quickly as possible,
The road subsided and cracked at intervals, especially near low lying lands.
Between the 20th and 21st miles of the road from Golaghat, it was noticed that
sand and water spouted up in places filling old earth borrow pits.

( 342 )

APPENDIX B. 343

Letter No. 292, dated Manipur, the 15th September 1897, from A. Porteous,
Esq., 1.C.S., Officiating Political Agent in Manipur and Superintendent of
the State, tothe Secretary to the Chief Commisstoner of Assam.

I have the honour in reply to your memorandum No. 5148-59 G., dated oth
August 1897, to state that, after enquiry, I cannot ascertain that there has been
any instance within the Manipur State of either earth fissures, outpourings of
sand or water, or of filling up of river channels having occurred as a consequence
of the earthquake of 12th June last.

Letter No. 3177, dated Silchar, the 2nd November 1897, from B. B. Newbould,
Esq., 1.C.S., Deputy Commissioner of Cachav, to the Secretary to the Chief
Commissioner of Assam. d

With reference to your memorandum No. 5148-59G., dated the oth August -
1897, asking for a special report giving information as to the extent to which
fissures in earth and landslips have been caused by the earthquake of 12th June
last, I have the honour to submit herewith a statement showing the information
wanted as precisely as is possible.

In Katigora tahsil the number of fissures and the area covered by outpourings
of sand and water were not recorded at thetime, and now they have been
obliterated. No river channels in this district were filled up by slips into the
river.

Letter No. 249G., dated Sylhet, the 7th April 1898, from L. $. Kershaw, Esq.,
I.C.S., Officiating Deputy Commissioner, Sylhet, to the Secretary to the Chief
Commissioner of Assam,

With reference to your No, 5148-59 G., of the 9th August 1897, calling for a
special report on certain matters in connection with the earthquake, I have the
honour to forward summary of the information I have been able to collect from
the various parts of the district.

2. The four heads under which the Director of the Geological Survey required
information are :—

(a) Extent of fissuring.

(b) Outpouring of sand and water.
(c) Filling up of river channels.
(d) Opening out of new khals.

As regards (a), fissuring has been universal throughout the district to a small
extent on the public roads which have been cracked rather than fissured, and to a
much larger extent on the banks of rivers and water courses and in the
neighbourhood of tanks, bils and haors. Fissures in the latter cases were caused
by the subsidence of the banks into the bed of the river or tank, and their direction
was, therefore, almost invariably parallel tothe banks. The extent of the fissures
varied, as might be expected, according to the distance from the centre of disturb-
ance. In Karimganj the breadth did not exceed 5 or 6 feet. In Habiganj and

( 343°)

P‏ یھ

344 OLDHAM: GREAT EARTHQUAKE OF 1897.

Sunamganj they seem to have been more considerable. In Sunamganj the
largest fissures were over 30 feet broad and extended along the river bank in
some places over half a mile. In Jaintia and Karimganj there are instances of
fissures extending in a continuous line right across narrow tongues of land, thus
nearly separating them from the main land and forming a sort of khal, fn
Jaintia along the banks of the river Surma the fissures in some places are 15 feet-
deep and usually over 6 feet deep. In some cases the bank which was cracked
at the time of the earthquake did not subside until November.

(b) The outpouring of sand seems to have been common everywhere in the
vicinity of rivers and bils. The damage due to this cause was not on the whole
very great, though inthe Jaintia Parganahs near Kanair Ghat in several
instances paddy was damaged by the upheaval of sand. The Subdivisional
Officer of Habiganj reports that the sand thus ejected has had a fertilising effect
contrary to expectation, and that the beds of tanks have interstices in the Dharma
pasa thana of Sunamganj. In Mouza Bayangpur, Parganah Satbank, Jaintia,a
khal was formed about $ mile long and was navigable for 3 malla boats! during
the rains. Two streams in Mouza Chikuagul, Parganah Panchbhag, have changed
their course. In Mouza Bugail Kandi, Parganah Dhargam, the Piyain has left
its old course and now flows through Bugail bil $ of a mile away from its old
bed. 5

Appendix C,

Details of observations made for the purpose of determining the
velocity of movement and acceleration of wave-particle with
a discussion of the formule employed and the results obtaining.

In this statement the values of f, or acceleration, is given in feet per second
and deduced by West's formula.

The value v is deduced from Prof. Haughton's formula as given in Mallet's
account of the Neapolitan Earthquake of 1857.

In the case of objects projected the value is obtained by the angle of projection
being taken at 45” or o” according to circumstances. Where the vertical difterence
of position much exceeds the horizontal the latter gives the minimum value,
where On the other hand the vertical difference is nothing or small, the former
gives the minimum value. |

[n every casethe figures in the last three columns are given in foot-second
units. These values may be converted into millimetres by multiplication by 300.
The error, which is little more than 1%, is immaterial in view of the much greater
errors inherent in the method of determination.

The second column states whether the value is a maximum or minimum one.
Thesign < representing a maximum value, or that the real value is less than the
given ; the sign > on the contrary indicating a minimum value, or that the real
value is greater than that given.

1 i.e., Beats requiring three boatmen.

(344419

345

APPENDIX C.

The first column explains itself, It is a brief description of the object utilised

Overthrow Projec-
and fracture. tion.
si v v
Es es So

112 56 ves
6:9
15°24 41 e
8:3 2°5 T
8:7 30 gen
147 46 Ae
ee $ 10°84
I3 4'0 See
10°3 4'1 ^
16:0 53 eee
12°5 4'6
e 8°75
12:8 4'0
1 Grp
; : 13:6
E ; 61
IO'S = T
15:3 6'7 ve
117 4۰0
12:3 48 T
I1'7 43
24 100

value,

VV

VVV

VVV VV V AAAV

VVAVV

real to given

and a reference to its detailed description.

©
NATURE OF OBJECT AND REFERENCE. 5
E
c
Dacca—
Top of pillar on race course £— 45? p. 291.
Race course pillar, not overturned p.318.
CoMILLa—
Top of temple projected e zo" p. 300 .
DHUBRI—
Telegraph office gate post . p.259.
Gauripur Zamindar's wall . p.317.
GOoALPARA—
Municipal boundary pillars . مم‎ A
Pillar near cemetery - : p.263.
Ensign Law's tomb ¿=30" . p.262.
GAUHATI—
D, C's compound pillars DILE Hr e
Judge's gate pillars , e D 263 .
Tel. signallers’ house, w, gate posts p. 264 .
Ditto ` ditto p.264.
Top of pillar in Civil Surgeon’s
compound e=45° 2 : p.264.
Gate pillars W, of D. Cs com-
pound . ٠ - : p. 265 .
Do. western gateway p.265.
Tomb of Anna Louisa Lamb, pro-
jected ` : p.265.
> R. Beecher, do. . p.317 .
Marble cross broken . : D, 317 .
SHILLONG—
Ashley Hall, western gate pillar p.270.

» eastern do. p. 270 .
Beauchamp lodge gate pillars p.270 .
Col. McGregor’s gate pillar . Rés:
Cylinder seismometer , ^ P.358 .

346 OLDHAM: GREAT EARTHQUAKE OF 1897.

5 3 Overthrow Projec-
“ba and fracture. tion»
NATURE OF OBJECT AND REFERENCE, Ee:
PEN رك‎
CHERRAPUNJI— : `
Inglis’ house gate pillars - p.954 < 83 35 2
e pillar on wall. 0101202 < 13'9 48 M
Hudson's house gate pillars . 70.٦ < 10'7 48 :
Cylindrical pillar . 5 e رت‎ < oi 36 a
Do. > . A 1و < 277م‎ 36 SEN
REMBRAI—
Stone projected £245? . ; pareri > oe 165
SYLHET= |
Tomb of Ensign Spooner ` p.274 . > 32 11'4 1
Gate pillar near hospital : p.275 - > 1173 4'0 +.
Church northern gate pillar . p.276 > 24 Sr ioe
» southern do. - P: 270. > 192 T9 ae
SILCHAR—
Cylinder seismometer . 21 053587 > 4 27 e
Bullet projected . : : p.258. > CR Ss 14
Gate pillar overturned . , p.318 . < 6:5 ir p

Remarks on the formule used and results obtained,

Taking the attempts to deduce the velocity or acceleration of wave-particle
from objects overthrown by the earthquake in order of date, we have first the
formula of Prof. Haughton, published in Mallets account of the Neapolitan
earthquake of 1857. *

The investigation proceeds as follows and may be given in Mr. Mallet’s own
words :—

“In order to upset the body, the
horizontal velocity impressed by the shock
(whatever be the duration of the latter)
must be sufficient to make it turn upon
one of the arrises or angles of its base
through an angle formed by the line,
Fig. 42, joining the centre of gravity
with that angle or arris, and the vertical
through it.

1 R, Mallet, Great Neapolitan earthquake of 1857. The First Principles of Observational
Seismology as developed in the report to the Royal Society of London of the expedition made by
Command of the Society into the interior of the Kingdom of Naples, to investigate the circum-
stances of the Great Earthquake of December 1857. 2 vols., London, 1862. Vol, ا‎ p. 125.

(340...)

APPENDIX C. 347

“ Let a denote the distance (in feet) of this point or edge from the centre
of gravity, then the statical work done in upsetting the body, whose weight
is W.
W a (1 — cos. ¢).
“This must equal the dynamical work acquired, which (as is well known) is
equal to the work stored up in the centre of gyration, or

W w? k?

28
where e is the angular velocity of the body at starting, & the radius of
gyration, with respect to the point or arris on which it turns, and g the velo-

city acquired by a falling body in one second of time.
* Equating these two values of the work done we find—

Wa(t—c0s.p) =

wb -—2ga (1 — cos. q)

but w, the angular velocity, is equal to the statical couple applied divided by
the moment of inertia, or

_ 9 a cos. 4

> p

squaring and substituting
nw E tpos $ >”
I= 2gp x — م‎ 1
4 sd "I cos” y (2)
For a pillar of rectangular section
re e
3

and a=} Vat 8

where a is the height and ĝ the side of the base; hence

)1 الہ یں مو رر
For a cylindrical column—‏
p 153? + 16 a?‏
and A‏
«roii UE‏
and since‏
a?‏
ES a+ (32‏
the equation (1) becomes‏
pT Ha‏
fa? + (os (12)‏ رو De‏

( 347 )

48 OLDHAM: GREAT EARTHQUAKE OF 1897.

The defect in theory underlying these equations is the supposition that the
impulse is instantaneous, it is assumed to act for an infinitely short time over an
infinitely short distance.

The next attempt to utilise overturned pillars and columns as seismometers
looked to them as measures not of the maximum velocity, but of the maximum
acceleration of the wave-particle. This is due to Mr. C. D. West, by whom
the argument is stated as follows! := ~

" Let the surface of the earth at any instant be undergoing an acceleration
of velocity of کر‎ feet per sec. per sec. Let
M be the mass of the column (see. fig. 43)
resting on the ground, y the height of its
centre of gravity, and x its horizontal
distance from the edge round which it
may be supposed to turn.

“Then the inertia of the column is
equivalent to a force

Fig. 43. Ez MF
acting horizontally through its centre of gravity and tending to overturn
the column, the overturning moment being
Fy=Mfy

“This moment is opposed by the moment of the weight of the column, W x, and
therefore when the column is on the point of overturning

Wc Mfr

7
I4
Acn (IT)

* [f f exceeds this value the column may go over, if less the column may

stand.”
This formula recognizes facts as they are more than (I) and in Prof. Milne’s

experimental investigations it gave results in somewhat close agreement with
the actual, but where the period was about 13 to 2 seconds, there were many wide
divergencies ; the calculated acceleration being in excess of the actual.?

The formula, it will be seen, takes no account of the amplitude, or of the
range of motion, t.e., double amplitude, of the wave-particle, and Prof. Omori
points out that the column will only overturn towards the direction from which the
impulse comes when its rotation is sufficient to bring the centre of gravity over the

edge on which it turns,’
1 Journ. Seismol. Soc., Japan, VIII, 35 (1885).

2 Seismol, Journ., Japan, XVII, 74 (1893).
3 Seismol. Journ., Japan, XVII, p. 120 (1893).

( 349. )

APPENDIX C. 349

But the impulse may be assumed to be imparted at the centre of percussion
P (fig. 44) whose distance from the base is

4 (y)
xs t
that is to say when
2a OA ;
OP^ 06 approximately

2 a being the double amplitude, or range of motion of the
edge A, hence

OA x o dei + y?)
2= xXx OP =~ X ك‎
SUE y 3y
وا نے‎ (x + y)
3317

In the same paper Prof. Omori suggests that, where the
amplitude is greater than this and the period a long one,
the pillar will, during the forward semiphase of the wave,
acquire a velocity of movement equal to that of the wave

Fig. 44. particle, and on the backward semiphase setting in, this

velocity may be regarded as having been suddenly imparted

at the centre of gravity of the pillar. Equating this with the work required to

turn the pillar till the centre of gravity comes over the edge on which it turns, he
gives, for a pillar of rectangular section—!

PAR‏ ےہ
0 3

JE o
3 cos? @

This formula, however, though different in form, is identical with Prof. Haughton’s
formula (I, 1) for the overthrow of a rectangular pillar.

or

Besides pillars that have been overthrown it will be found that others have
been broken across but not overthrown. For determining the velocity of waves
particle from this, only one formula has been proposed, though its form mzy
be modified. This is the one given by Mallet—*

FLA k?
mA VI URB
where

W is the weight of the mass broken off

! Journ. Seismol. Soc., Japan, XVIII, 121 (1893).
2 Great Neapolitan Earthquake, I. 141.

( 349 )

350 OLDHAM: GREAT EARTHQUAKE OF 1897.

h. The perpendicular height of the centre of gravity above the base of fracture.
F, The coefficient of dynamic cohesion, or the force upon the unit of surface
of the material fractured, which, when suddenly applied, is sufficient to
produce fracture.
A. The area of the fracture in such units.
k. The radius of gyration of the plane of fracture round the edge.
6. The width of the fractured surface.
In practice the value of F, cannot be directly determined, but as it is taken as
half of the breaking strain under a load gradually applied, which may be
represented by F, the formula becomes

or substituting 2 x for ff and y for hto bring the symbols into uniformity with

previous formuiz
FA

a الع‎ 2

& Sy Wy x Xk

<
Il

or, since
W=2y Aw

where w is the weight per unit volume,

F
= یچ ھت‎ 2
4 EX Sway =

— L 2
=S Bx ue e x
Where L is the modulus of cohesion, or length of a column whose weight is

just sufficient to cause fracture.
It must be noticed that this ٣ is not wor the maximum velocity of wave-
particle, but the velocity which if suddenly applied would cause fracture, or,

in other words the acceleration, hence we may write the formula

FA ` F Y,

= ھی کے کشر لت‎ re D
LE A Wy x E (Suzy e 887 y
For a wall or rectangular column
تع‎ ee Mus
3 3
hence
E E E E
For a solid cylinder
P= D = A a?
hence
NUS TA. 5 Fea 5 7 & Sa?
f = EE E “gy Gv)

( 850.3

APPENDIX C. 351

There is yet another means of determining v or the maximum velocity of `
movement of wave-particle, and that is by means of bodies which have been
projected through the air by the earthquake. If we knew the exact angle
at which they started on their course it would be easy to calculate the velocity
with which they started from the equation

GN سو و شف یر و‎ (Y)
2 cos? e(b + a tan. 8)

where a and b are the horizontal and vertical distances through which the centre
of gravity has moved (the latter downwards) and e the angle with the horizon at
which it was projected upwards,

In practice € cannot be determined, and the only course to pursue is to
adopt a value which will give a minimum value for v. Where û is very small as
compared with a, e may be taken as 45° and the equation becomes

(€ = 45°) i r 0ا‎ (V 1)

which becomes, when 5 is nothing, as in the case of projection on level ground,

v=ag
Where b is much greater than a, or the drop is much greater than the horizontal
distance travelled, then ¢ may be taken as o”, that is to say, the projection is
horizontal, and the formula becomes

Le = 07) pz EL (V 2)

Of these formulz, that for deducing the acceleration from fracture has
not been used, except in the case of the marble cross in the Gauhati cemetery.
In the case of the broken tombstones and Khasia monuments it could not be
applied owing to the impossibility of determining value of either F or L, nor would
it have been any use to test the strength of pieces of stone taken from the
monuments, as the irregularity of the weathering the stone has undergone would
make it impossible to assert more than that, whatever the breaking strain of the
piece tested, that of the stone on the fractured surface would almost certainly be
different. Similarly the irregularity and general badness of the brickwork
in Assam rendered it useless to attempt to determine the breaking strain, as the
strength along different joints, in the same pillar, obviously varied very much.

l Generally speaking, it seemed as if the strength was such that the acceleration

required to produce fracture was about the same as that required for overthrow,
for, of the cases where one of two similar pillars were overthrown and the other
not, there were about an equa] number where the standing pillar was cracked
across and where it was uninjured.

Referring to the tabular statement at the head of this appendix it will be seen
that the values obtained for the acceleration are fairly concordant on the whole.
As exceptions must be noticed the cylinder séismometer at Shillong, the tomb

[ E,

352 OLDHAM: GREAT EARIHQUAKE OF 1897. :
of Ensign Spooner at Sylhet, and the gate pillars of the Church at the same
place. In the case of such very squat pillars, or overthrown portions of pillars,
as these, it is probable that overthrow is not due to high horizontal acceleration,
buttoalarge vertical component of the motion. They must in fact be looked
upon as modified cases of projection, and it may be doubted whether. the formula
can be depended on in those cases where the proportions of the object overthrown
are such as to give accelerations of over about 12 feet per sec. per sec.

It will be seen that at Gauhati these are a number of cases of overthrow giving
values of f ranging up to 17 ft. per sec. per sec. Excluding the high values
as suggested in the last paragraph, we get an average acceleration of about
12 Ít. per sec. per sec. while the objects projected by the earthquake (excluding
the tomb of Mrs. Lamb, where the projected pillar seems to have rolled after
touching the ground) we have a velocity wave particle of from 6 ft. to 83 ft.
per sec.

If the motion of the wave-particle was a simple harmonic one, as it would be ina
purely elastic wave, and as it has generally been supposed that it may be assumed
to have been, without any material error, in the case of an earthquake, it would
be possible to calculate, from these data, both the amplitude and period of the wave
motion.! Butif we attempt to combine the acceleration and velocity deduced
we obtain a wave period of about 4 seconds, and an amplitude of about 4 feet,
both results which are preposterously impossible.

The difficulty is somewhat diminished, though not removed, if we remember
that the tabulated values of the maximum acceleration are those of the horizontal
component only, and that the fact of stones having been projected upwards shows
that the vertical component of the acceleration must have been greater than that
of gravity. Taking it as only equal, that is at 32 ft. per sec. per sec., and the
horizontal component as 12 ft. per sec. per sec. we get the resultant total acceler-
ation as at least 34 ft. per sec. per sec. Combining this with a maximum velocity
of 8 ft. per sec. we find that, on the supposition that this velocity and acceleration
were due to a purely elastic vibration, the amplitude would have been about 2 ft.
and the period 13 sec., while if the maximum velocity was only 6 ft. per sec. the
amplitude would be t ft. and the period 1 sec.

Here there is nothing to find fault with in the deduced period of the wave, but
the amplitudes, involving a total range of motion of 4 feet and 2 feet, respectively,
are far beyond what there is any reason to believe took place; and it must be
remembered that the maximum acceleration and the maximum velocity of wave-
particle were certainly not less than the amounts utilized in making this calculation,
but were probably greater.

1 From the formula for simple harmonic motion

٤ J= as‏ و سد رہ
t : a‏
We obtain‏
q? 2 Tn‏

APPENDIX C. | 353

The difficulty is far greater when we consider the case of the projected stones
near Rambrai. Taking the maximum velocity at 16 ft. per sec. and the ampli-
tude as 1 foot, which may be regarded as certainly in excess of the amplitude of
the wave motion, we obtain the incredible result that the maximum acceleration
was 256 feet per sec. per sec. In other words the violence of the shock was equi-
valent to that of sudden arrest after a fall from a height of 1024 feet. Even if we
take the impossible amplitude of 2 feet, involving a backward and forward move-
ment of double this, or 4 feet, the acceleration is only reduced to 128 feet per sec.
per sec, or equivalent to sudden arrest after a fall of 256 feet.

It must be clearly understood that, if the movement was due to elastic vibra-
tions or to a wave motion even approximately simulating such, there can be no
mistake in these figures. The stone was projected through the air a distance of
84 ft. horizontally and to enable it to traverse this distance the initial velocity must
have been at least 163 ft. per sec., and if the direction in which it was projected was
at an angle much greater, or much less, than 45° to the horizontal the velocity must
have been even greater.

Knowing the maximum velocity of wave particle, and the amplitude of the
wave motion, the maximum acceleration can be infallibly calculated ; but we have
seen that even assuming animpossibly great amplitude we obtain an incredible
acceleration. Where then is the error ? Evidently in the assumption that the
high velocities determined were those of wave motion.

The results of the revision of the great triangulation in the Khasi hills show
that there have been considerable changes, both horizontal and vertical, in
the positions of the stations; yet the chain of triangles lay near the outskirts of
the epicentral tract, where displacements might be expected to be at their mini-
mum, The observations recorded in Chapter IX show that there have been
differential movements approaching 40 feet in vertical height in parts of the
epicentral tract, and it is conceivable that in places the actual changes in height
and position have been far beyond those which have been measured as yet. This
being so we would have, in the epicentral tract, bodily displacement of the ground,
apart from true wave motion, which would probably take place by a series of
jerks, but might have been a simple and[single movement.

Totakethe case of Rambrai, if we assume that there was a permanent
displacement of the ground ef only 16 feet, and that the movement was approxi-
mately harmonic in its character, Ze, beginning slowly, gradually increasing in
rapidity and then gradually dying out, then the required velocity of 16 feet per
sec. would only involve a maximum acceleration of 32 feet per sec. per sec. That
the acceleration was at least this much we know from the fact that, in order to
account for the projection of this block of stone, the vertical acceleration must
have been at least 32 feet per sec. per sec. and the maximum acceleration in the
ground consequently greater than this.

From this it will be seen that the high velocities of movement which were de-
veloped in the epicentral tract cannot be ascribed to the wave motion proper, set
up by the earthquake, but must be attributed to those bodily displacements in the
earth's crust which were the primary cause of the earthquake. In other words,

aA ( 353 )

354 OLDHAM: GREAT EARTHQUAKE OF 1897.

there were, within the epicentral tract, great movements of permanent displace-
ment, superadded to the molecular vibratory movement of the more or less purely
elastic waves set up by the earthquake; and it is to the former of these, rather

than to the latter, that high velocities of projection within the epicentral tract
must be attributed.

Appendix D,
Details of the direction of overthrow and projection of free fall-
ing objects, and a consideration of the applicability of this form
of evidence to the determination of the position of the epicentre.

In the following list are given all the cases that have been observed of objects
projected or overthrown. In the case of objects overthrown by the earthquake
only those whose direction of fall was not largely controlled by their form have
been included. Such may be designated as “ free-falling,” and are almost exclu-
sively square or round pillars or pinnacles ; objects having an oblong section with
One axis much longer than the other have been excluded, as they were not free to
fallin any[direction, but must necessarily fall more or less at right angles to
their longer axis.

Overthrow j
NATURE OF OBJECT AND REFERENCE, or ` Azimuth.
projection.

CALCUTTA ٠ 3
Sir W. Jones’s Monument - : : ps 257 O 200
Mrs. Wiltshire’s ditto e . . Ps ص۵‎ O 207
St. Paul’s Cathedral spire. . : - p- 258 O 2113
Town Hall balustrade : - e : ps) 258 D 200
Free Kirk spire . e . TPs 209 O 200

JALPAIGURI— ` P 2
Cupolas'on District Board's Office . e Pa 52819; P 3573

DARJILING— :

Bhutia tombs . . . . e ٠ p. ٤ 0 128
Chimneys at Senchal. . . 1B. 293 O 128
RANGPUR—
Images on gate-posts of Dimla Rajah'shouse p. 285. O 965
Tombs in cemetery S 5 7 5 p. 285 0 97
O, t . D D 7 . D 285 0 277
Pillars of Library verandah e BRR 285 O 943
Gate-post of Collector's house . o . p. 2905. O 943

——M LL TEL =

ÉL xit)

APPENDIX D. i 455:

Overthrow
NATURE OF OBJECT AND REFERENCE. or, Azimuth.
projection.
Kucu BIHAR—
Marble vase in Palace . . . . p.288 . 0 IIO
Clock Tower . : p. 288 0 108
Urn on turret of Superintendent's house . p.288 . F 110
Gate-post of Superintendent's house ء۲٦‎ O 1024
Watch Tower > e : : < PBa . O 108
Minaret of mosque ° . : : o رد‎ O 298
Buxa Duars—
Prevailing direction : : : ^ p. 288 O 139
GOBARDANGA—
Turret on Zemindar's house e e . Pp. 315 +. O 205
FARIDPUR—
Bottles everturned : d d s à p.317 + O 29:
Dacca—
Pillar by Race course A z p.290 . ۲ 115
Nazir’s Mutt , R ut 3 3 + — p.291 O 335
MAIMANSINGH—
Hindu temple : A ae «293 . P 48
DHUBRI—
Telegraph Office gate mm ا‎ s TEC Lo We 0 40
Gate pillars : e . . p.260. 0 o
Gate pillar : į 1 e : ا‎ Cup ds 0 355
O. A e o B o " p.317 + 0 145
GOALPARA—
Tubs B $ e e o P: 262 . P 357
Tomb of Ensign Law . . . . p.262 . P 355
Pillar near cemetery e . ٠ = p.268 . 0 120
Boundary pillar No. 1 e e e $5 ep, aig O 50
Do. 2 : ^ A : p.319. O 273
Do. 3 d Les پا‎ AT 0 63
Do. 4 e sg e e P. 317 0 295
Do, 5 . ٠ e METE DU O 235
GAUHATI—
Deputy Contmissioner's Bungalow pillars p. 264 . O 150
Do. : : cu oq 262 « O 140
Telegraph Signalter” 5 quarters : p.264. O 200
Top of pillar in Civil Surgeon's Compound p. 264 . E 145
Compound, west of Deputy Commissioner’s
bungalow. ۰ E R S p.265. 0 200

2A2 | ( 365 )

350 OLDHAM: GREAT EARTHQUAKE OF 1897.

ee EEG A A AA AAA AAA AAA A ETS ES SS ہے کے‎

Overthrow
NATURE OF OBJECT AND REFERENCE, Or Azimuth,
e projection.
GauBaTi—contd.
Compound, west of Deputy Commissioner’s
bungalow e H e e e 0 265 O 290
Do. e e D D D, 265 O 325
Do. e e * . P 265 D 0 270
Do. a EE O 100
Do. e e. e e P 265 . O o
7 d ‘ : e 2 pear O 340
Tomb of G. T. Bayfield . : e : p. 265 O 160
F, Gomes D e e e p 265 A O 155
Cylindrical tombs NE و تل‎ nens E 315
Do. : . : : e «x ID. 2OS ne O 225
Miss Lamb : Š : S Mg te ۲ 180
Loki Rani’s house, gate pilan par. S : وحم‎ 0 60
Gate pillar near Telegraph Club . e p.317 O 167
Do. : ` 5 0۰ 0 140
Monument to Robert Beecher + : 27 ۲ 165
CHHATAK—
Gate Pillar of Monument to Lieutenant H.} p. 218 O o
T. Bird Gest ES ` P. 318 O 180
SBILLONG—
Gate pillars e e . . e p.270 + O 180
Ashley Hall gate post. +... Bän, O o
Do. e 2 e 1 DU 0 180
Colonel Magregor’s gate post . ^ —. مڑوھو‎ O 350
ther gate pillars ٠ . a . p.279 . O 175
Do. e e o 6 . e p. 270 O 205
Do. : » owe - 42 po. O 180
C HERRAPUNJI—
Inglis’ house gate pillars . e e ET 0 200
Do. . : > EE 0 135
Gate pillar . S : p.273 + O 50
SS = Bungalow « on East road 5 - p.318 O 257
Do x» - e ٠ p. 318 O 40
» Dak bungalow . : . Polk O 210
Do. e D D D D D. 318 E 0 205
Hudson’s bungalow gate pillars . 318ص‎ O 230
0. $ z : > p.318 . O 235
Verandah pillar : : 2 E p.318 A OT 23 3
Do. . o . . 0 e D, 318 . 0 55
Do. e : - : ? ' p.319. 0 153
Do. ۰ D e . e D 0 318 ۰ 0 75
SILCHAR—
Cylinder Seismometer - e ` ; p.359 . B 4
o

Gate pillar, Mission house S SC EEN DS S

0ئ ۴)

APPENDIX D. 357

lO کے‎ e ee ae e e
A

Overthrow
NATURE OF OBJECT AND REFERENCE., or. Azimuth,
projection.
SYLHET—
Gate pillar near Government School . - p.274 . O 40
» Deputy Commissioner’s bungalow p. 274 + O o
Tomb i in cemetery e e / - p.274. o O 155
» of Ensign Spooner . . - . P.275 » O go
Brick pillar e : . ` : . Pp. 275). O 180
- Square column . - - e p. 275%, 0 225
۔‎ Gate pillar, New Hoa! a e - P270 O 283
.Chseehe e e : e p.220 , O 0
Do. e : , e p.276. O 250
Vertical boiler ا‎ e : : . p. 276 £ O 240

In the early days of the modern development of seismology, great importance
was attached to the direction of overthrow of free falling objects, as a means of
determining the position of the epicentre and hence of the focus. This method was
employed by Mr. R. Mallet, in the case of the Neapolitan earthquake of 1857
with a concordance in the intersection of the lines of direction obtained, so
conspicuous as to give great support to the supposition on which the method was
based. In the case of subsequent earthquakes, however, the method has not been
found so successful, and with the growth of amore accurate knowledge of the
true nature of earthquake motion, as derived from the labours of the band of
workers in Japan who have practically revolutionised seismology, the method has
become discredited.

It was found that instead of being a simple to and fro motion, the motion of
the wave particle was extremely complicated, and that, in the case of moderate
earthquakes, the maximum movement was by no means always or even usually
in the direction of the propagation of the earthquake wave, but might be at any
angle with it. Assuming that the same held good of more violent earthquakes
it was evident that the direction of overthrow would only exceptionally be that
of the propagation of the earthquake wave, and would consequently generally
point in any direction except that of the seismic vertical.

Against this there has always stood the case of the Neapolitan earthquake
where the direction- of overthrow in the vast majority of cases pointed to the
epicentre, and it has long seemed to the writer that the assumption, that the
nature of the movement in violent and in moderate earthquakes was similar, went
too far. It is conceivable that, within the limits where the shock has still power to
cause the overthrow of gate pillars, etc., the one or two vibrations, which alone have
this power, should be directly or nearly directly outwards from the centre, while
further away this relationship might cease to hold good.

An examination of the tabulated directions. of overthrow and projection lends
some support to this supposition. At places like Shillong, which were within the
limits of the epicentre, or, like Gauhati, which were close to it, there is no uniformity

[ 0

358 OLDHAM: GREAT EARTHQUAKE OF 1897.

in the direction of everthrow. At a greater distance from the epicentre and
where it has come to subtend but a small angle, it is to be remarked that the
directions of overthrow almost all point towards it,

From this it would seem that the method of determining the epicentre by
means of the direction of overthrow of free-falling objects is one that may be
depended on, when other means fail. Certain obvious precautions must be taken.
The objects whose direction of overthrow is taken must be of such a shape that
they have no greatly prevalent tendency to fall in one direction rather than another
and they must have been overthrown cleanly. If the surface of fracture shows
great crushing at the edges, indicating that the pillar has rocked to and fro before
falling, it should not be utilised, as the final direction of overthrew is almost cere
tain not to be that in which it was first set rocking. Another precaution to be
observed is that the observations must be sufficiently numerous, and the inter-
sections lie pretty close together. This is necessary for, though it is certain that in
some earthquakes the greatest and most violent motion is directly inthe line of
propagation of the earthquake wave, it is possible that in others the reverse may
be the case, and, if so, the directions of overthrow will not, if produced, intersect at
or near the right point.

Appendix E.

Details of the records of the Seismometers at Shillong and
Silchar.
In 1882 seismometers of Mallet’s cylinder pattern were set up in Shillong

and Silchar. Each consists of aset of cylinders 12 inches in height and of the
following diameters :—

No.o P No. 3 3°75" No. 6 gel
جو‎ I 1°5” 5 4 50 a ef 8:0"
رو‎ 2 2'5” 35 5 Go وو‎ 8 oo
The corresponding accelerations in ft. per sec. per sec. required for overthrow
are:—
No. o 217 No. 3 10'0 No. 6 18:8
» i 40 D A 130 وو‎ 7 215
» 2 67 وو 161 5 وو‎ 8 242

Besides these there was at Silchar a square wooden post, 4'o5 ft. in height, at
esch top corner of which a cubical recess was cut, containing a small bullet
to be projected by the shock. Full details of these instruments will be found in

Memoirs Vol, XIX, Pt. 1, Appendix, pages 93*c8, and need not be repeated here,

Shillong,—At Shillong the seismometer shed is close to the hospital, and was
the only building left standing after the earthquake. It was used by the hospital
assistant to shelter the patients, and all that is known of the effect of the earth-
quake is that all nine cylinders were upset in a north-north-easterly direction.
(The cylinders were not reerected for three days, during which several shocks

( 358 )

APPENDIX E. 359

otcurred sufficiently violent to overthrow some of the cylinders. For the subsequent
records 1 am indebted to Surgeon-Lieutenant-Colonel R. Neil Campbell, Civil
Surgeon of Shillong.

12th June . 5-15 P.M, -
9 cylinders fell, south-south-east to north-north-east.
16th June + during night, time not noted.

1 cylinder fell, south-west to north-east.
19th June , 5°15 P.M,

3 cylinders fell, south-west to north-west,

Ditto » 10-30 P.M.
2 cylinders fell, south-west to north-east.
e during night, time not noted,

1 cylinder fell, north-west to south-east.
12th July e 8-35 P.M. -

2 cylinders fell, west-north-west to east-south-east,
2nd August . 9-30 P.M.

3 cylinders fell, Nos, o and 2, west to east.

No. 1 south-south-west to east-north-east.

The shock of 5-15 P.M. on 19th June is a remarkable one as, in spite of its
severity, it seems to have been local; it was felt at Tura and, apparently, in
Gauhati, but elsewhere I have not been able to identify it in the records received.

Stlchay.—Here 2 cylinders were overthrown, No. o to N 4? E, No. 1
to N 43? E ; of the spheres the one in the north-east corner was thrown 4 inches
to E 10? S; the south-east corner, 81 inches to S 47° W ; the south-west corner
3i inches to S 69° W; while the bullet in the north-west corner was not dis-
placed.

No subsequent shock affected the seismometer at Silchar.

These details were communicated by Surgcon-Lieutenant-Colonel S, Borah,
Civil Surgeon.

Appendix F.

Details of the special observations made by the Telegraph Depart-
ment, referred to in chapter X.

These details are extracted from the reports of the telegraph masters who were
instructed to observe the intervals of time between the arrival of shocks at one
station and its neighbouring ones. The name of the station in italics is that
of the reporting one and what follows an abstract of the report :—

September 27th 12-29 Shillong, severe; felt at Sylhet simultaneously, but
very slightly.
21-38 Shillong, severe and prolonged ; felt by Gauhati simul-
taneously, very severely.
> goth 19°46 Shillong, severe; felt by Gauhati about 1 minute
before us.
October 3rd 7-0 Gauhati, Goalpara and Dhubri, felt a severe shock
not noticed at Shillong.

( 359 )

360 OLDHAM: GREAT EARTHQUAKE OF 1897.

October sth 9°55 Shillong, two slight shocks followed by a very severe
one ; felt severely at the same time by Gauhati.

6th 4*1 Shillong, severe; felt by Gauhati at the same time.

7th ro-22 Shillong, severe; felt slightly by Gauhati at the same
time.

oth I-7 Shillong, very severe; felt at Gauhati severely, simul-
taneously, not noticed by Sylhet. Dhubri, reports the
same shock, message sent to Gauhati and noreply
for 30s.; (this appears to be due to a delay in reply-
ing and not to a true interval of 30s. between the
shock being felt at Dhubri and Gauhati).

11th 12-45 Shillong, smart; felt at the same time by Gauhati
slightly; not noticed by Sylhet.

13th) 1444 Shillong; very severe; felt by Gauhati 2s. after us,
whence it was reported as ‘smart’; not noticed by
Sylhet.

» 14th 8-19 Shillong, severe; felt by 6 25. before us; Gau-
hati reports it as ‘ severe.’

18th 10-35 Shillong, severe ; felt by Gauhati 2s. after us, ‘very
severe’; by Sylhet simultaneously but slight ; original
signal sent from Sylhet. Dhubri reports ihe same
shock, a signal sent to Gauhati was replied to imme-
diately.

21st 13-46 Shillong, severe; felt 3s. Or 4s. after Gauhati; not felt
by Sylhet.

22nd 937 Shillong, severe; felt at same time with Gauhati.

15-32 Shillong, severe ; felt by Gauhati and Syihet at same
time; Gazuhati sent original signal ; 00001 felt
the shock at the same time.

25th 48۔5‎ Shillong, slight, felt by Gauhati same time, not felt
by Sylhet; Gauhati, lasted Goes, original signal re-
ceived from Shillong, time taken for earthquake to
reach Gauhati one minute from time of original signal.

27th 12-31 Shillong, a slight shock followed by a smart one ; felt
by Gauhati in the same way; but 5s. after us ; not felt
by Sylhet.

»» 31st 8-55 Shillong, severe; felt at the same time at Gauhati.

November 2nd 4-31 Shillong, severe ; felt slightly by Gauhati at the same
time.

10-55 Shillomg, severe; felt at the same time by Gauhati.
Gauhati original signal from Shillong, interval
between shock felt at Shillong and Gauhati 1 m.

2nd 11-48 * Shillong, slight, felt by Gauhati 5s. before us.
Gauhati, simultaneous shock at both places.
13th 13-11 Shillong, smart, signal received from Gauhati 2s.
before shock.

33

7

23

>

APPENDIX 7 361

November 15th 4-50 Shillong, severe ; felt tos. after Gauhati.
17th 20-17 Shillong, very severe, felt by Gauhati and Sylhet at
the same time. ,
F 26th 9-50 Shillong, severe, felt by us about 2s. after Gauhati.
December 17th 17-42 Gauhati, felt 3s. before Shillong.

”

Appendix G.
Details and Discussion of the results of the Revisionary Triangu-
lation `of the Khasi Hills, carried out by the Great Trigono-
metrical Survey in the season 1897-96. |

The accumulation of independent evidence, from many witnesses and widely
separated places, all pointing to the fact that the earthquake had been accom-
panied by noticeable changes in the relative heights of peaksin the Khasi and Garo
Hills, led the Government of Assam and the Director of the Geological Survey
almost simultaneously to represent to the Government of India the desirability
of testing the accuracy of these statements by a revision of the triangulation
of the hills. The representations met with a favourable reception; it was
decided that retriangulation should be taken in hand during the cold weather
of 1897, and as much done as was compatible with the other calls on the staff of the
Great Trigonometrical Survey. In pursuance of this decision, a programme of
work to be done was drawn up by Major S. G. Burrard, R.E., Officiating Super.
intendent of Trigonometrical Surveys, in consultation with myself.

At the outset it became evident that what was desirable was not possible, owing
to the calls on the Trigonometrical Survey due to the total solar eclipse, and to tbe
regular work wl ich could not be interrupted. Only one observer could be deputed
for the revisionary triangulation, and the only instrument available, though an ex-
cellent one, was not of the power considered necessary for primary triangulation.
In these circumstances a compromise was made between what was desirable and
what was possible, and the decision come to was to reobserve certain triangles in
the Khasi hills, and, if time allowed, certain others in the western Garo Hills. This
latter part of the programme it was not possible to carry out, and the triangles
actually observed were those represented on Plate XLIII.

In coming to this decision it was recognised that a triangulation so much in the
air, and unconnected with any points which could with any degree of certainty be
regarded as undisturbed, would give no information as to absolute displacements,
whether vertical or horizontal. All that was looked for was a discovery of whether
there had been any relative displacements of the trigonometrical stations ; if none
such were found, it would be certain that no general displacement, which would
be detected, had taken place. With such an object in view it was obviously ne-
cessary that the triangles selected for reobservation should be situated where there
was the greatest inherent probability of their having undergone distortion. At the
time the decision was come to, the exploration of the epicentral region, described in
Chapter IX, had not been commenced, and we were still under the impression that
the focus most probably lay under the southern part of the Khasi Hills,

[301 g

e

352 OLDHAM: GREAT EARTHQUAKE OF 1897.

Another principle that was accepted as a guiding one, was that there was no
use in reobserving what are known as Tower Stations ; that is stations where, inthe
absence of natural elevations, brick towers had been built on the alluvium, from
which to observe. If retriangulation was necessary for geodetic purposes, all these
stations would require refixing, for within the area over which earth fissures were
formed, it would be o i to assume that any of them had not been disturbed
by displacements of the alluvium onewhich they rested, and this quite independent
of any change or absence of change in the underlying rock,

A group of triangles extending northwards from the southern margin of the
Khasi Hills was accordingly selected, and as there happened to be an isolated hill
in the Sylhet plain, Taramun Tila, which had been selected as one of the points
the original primary triangulation, it was decided to include this.

These preliminaries having been arranged the work wastaken in hand by
Mr. J. Bond, Extra Assistant Superintendent, and the results have been embodied
in a report by Major Burrard, from which the following tabular statements of the
changes observed are taken.

Changes in length of sides.

Side. 1860 value. 1898 value. Difference.

| Feet Feet. Feet.
Taramun Tila—Mopen ; ` d : 99101'5 ` 991063 +48
Mopen—Rangsanobo , . . e . 05013*1 9500 +4'9
Rangsanobo—Thanjinath : e - : 60421'3 . 604247 +24
Mopen—Mosingi . : - > 5 > 65344'6 66 oo
Rangsanobo—Mosingi . . . . 5 541525 ` 541548 +3
Rangsanobo— Mun e e ٠ . e 67892°7 67895'9 +3'2
Thanjinath—Mun . e - - . > 47884: 1 47884'0 —O0'1
Mosingi—Mun E E e . - 8482 84897'6 +4°4
Mosingi—Mautherrichan e : . " 83931:8 83933'0 T1'2
Mosingi—Laidera . - ^ S - : 64350"8 64353'4 +2'6
Mun—Laidera . e E e : . 630077 63007'4 —0'3
Mun—Dinghei . e 4 e e . 721543 72156'7 +24
Laidera—Mautherrichan . . . . 72373'1 723770 +39
Laidera—Dinghei . S : P S e 570 575841 +1'1
Mautherrichan—Landau Modo ~ - : 50746'7 50755'3 +6
Dinghei—Umter . . e e e el 0 7 3 4 2°3

( 362 )

APPENDIX G. 363

Changes in heights of Stations.

STATION, 1860 value. 1898 value. Difference.

Taramun Tila E Ce ET I 144 150 +6
Mopen . : - ` ٠ : e . 2,581 2,577 و سے‎
Rangsanobo . . . e : e e sep -e SE

RE lA :; 4,440 4,443 +3
a A eebe? 5,794 5,798 | + 4
Mun . e ٠ . e . e : 6,212 6,214 + 2
Laidera e gu ds SUE پور‎ a 6,180 6,186 +8
Mautherrichan . ٠ e e . e 6,288 6,312 +24
Dinghei > E A - - . 6,067 6,074 T
Landau Modo - e ‘ A ` . 5,160 5,177 +17
A Ke gt eC a Ng 37 | 3,670 Tg

Displacement of Stations.

LATITUDE. LONGITUDE. TOTAL,
STATION. ۱
Amount. | Direction. | Amount. | Direction. | Amount. | Direction,
Ft. Ft. Ft,
Mopen . ~ e è 2 N 5 5 NW
Thanjinath . A ` 6 N 2 E 6 N
Mosingi d : e o F 3 3 W
Mun ۰ . . " 4 N 0 4 N
Eamets’ ار‎ EE 2 N o 2 2 N
Mautherrichan S ‘ 1 N 5 W 5 W
Dinge ti ۱۷ 8 N 4 W 9 NW
Landau Modo . . 9 N 7 W 12 NW
Umter . . . . 7 N 8 Ww 11 NW

The figures in these tables refer only to the principal stations ; besides them
I am indebted to Mr. Bond for the following information regarding the displace-

( 593 ]

364 OLDHAM: GREAT EARTHQUAKE OF 1897.

ments of other stations, which seems worth publication, though the data were less
complete and the calculations less rigorous than in the case of those printed in
Major Burrard’s report.

Changes in height and position of Secondary Stations.

ALTERATION IN

STATION, Height. i Position.
Ft, Ft. Direction.
Suair - . . e . . " . - + 2'6 SCH N
Rableng . e " . SE . . . + 7 71 N
Shillong . . e ۰ . e e . . + 42 7'1 N
Somullon . - . . . . . SO 10°3 N
Laitbli . - . e - ٠ . - - + 90 2:1 NE
Mairang . . . e . . . و رگ‎ + 8 H 2
Kollong Rock . . e . . . . 3 T 14'4 5'6 W
Sniang . . . e . . e : ? 9'2 W

ص

A cursory inspection of these tables shows that there have been differential
movements of the stations, with respect to each other, which are too large and too
irregular to allow of their being attributed to errors of observation. The tabulated
displacements appear also, at first sight, to indicate that there has been a general
elevation and extension of the hills, such as might follow on a bulging upwards
of the surface due to the intrusion of a large mass of molten matter underground.
A more detailed examination, as will be shown, does not bear out this conclusion.

In starting any calculation, it was necessary to assume one side of one triangle
unaltered and accept this as a base; but it is clear that within the area over
which the triangulation spread no single side was left unaltered and there is
consequently no fixed base to startfrom. To quote Major Burrard’s report.—

“ The above results go to show that all Mr. Bond’s triangulation lay within
the area affected by the earthquake: no reliable base is consequently
forthcoming for the computation of his observation, Whatever side or
height we start from, we find that all other sides and heights have altered,
but the errors of our initial side and height pervade the calculations, and
the movements caused by the earthquake are given differently whenever
we change our basis.

( 364 )

APPENDIX G. 365

For the results tabulated above the side Rangsanobo—Taramun Tila was
adopted as the initial base for the triangulation, and the height of
Rangsanobo as the initial height, Mr. Oldham being of opinion that
these southerly stations are least likely to have been affected."

On which I need only remark that I am still of the same opinion, but consider
that it should have led to the adoption of Taramun Tila rather than Rangsanobo
as the starting point to which changes of elevation should be referred.

I had suggested that, as compression in a north-south direction was to be
expected, while compression in an east-west direction was less likely to be met
with, the assumed fixed base should be a side of a triangle bearing about east and
west. In the final calculations the side Rangsanobo-Taramun Tila was adopted
as the base, but whether from a desire to adopt a base of which one extremity
might be considered as unchanged, or for convenience of calculation, or because
it gives the smallest amount of general displacement, there is nothing to show.
What now remains to be investigated is the probability of any change having
taken place in the length of this side, and the effects such change would have on
the deduced values of other stations.

Taramun Tila is a small hill of tertiary beds, rising out of the Sylhet plain
near Chhatak. It lies outside what I believe to be the limits of the epifocal
area and,as it is a hill station, is not liable to the displacement which would
probably have affected a station built on the alluvium in this neighbourhood.
The position and height of Taramun Tila may consequently be regarded, with
considerable probability, as unaltered.

Rangsanobo, on the other hand, lies near the southern edge of the Khasi Hills,
and apparently within the limits of the epifocal area. In this case a displacement
is probable, and this would probably be such as to cause a shortening of the
distance between it and Taramun Tila. Consequently it is probable that the
present length of this side is shorter than of old.

The effect of this shortening would beto increase the apparent length of all
other sides, for, the base being assumed to be greater than it really is, this would
lead to the other sides being calculated as larger than their true length, in a
corresponding degree. Moreover, as the error would be cumulative it would lead
to just such a stretching out of the ground as seems to be indicated by the
tabulated results.

The supposition that the apparent stretching of the hills is really due to the
adoption of too high a value of the initial base, is confirmed on a more detailed
examination of the tabulated changes.

It will be noticed that the sides Mopen—Rangsanobo—Thanjinath, which run
nearly east and west, have apparently lengthened 4'9 feet and 3'4 feet, respectively.
Connecting these three stations with Mosingi and Mun, we have four sides with a
considerable amount of northerly departure in them. Of these one appears to
have shortened 1 foot, another is unchanged, and the remaining two have
lengthened 2'3 and 3:2 feet, respectively. Here the increased length which should
have been found if the shape of the triangles had been unaltered, is greatly

( 365 )

366 OLDHAM: GREAT EARTHQUAKE GF 1897.

diminished or has disappeared; this may be attributed to a compression of the
hills in a meridional direction, for the next east-west side, that joining Mosingi.
to Mun, again shows an extension of 4'4 feet. North of Mosingi and Mauther-
richan the sides again exhibit a marked diminution of the apparent expansion, but
the distance between Mautherrichan and Laidera has increased 3'9 feet, in a
W. by N. direction. Beyond this the changes get very irregular, and it is
impossible to say how far they may be.due to change in the assumed fixed base.

The conclusion pointed to in the last two paragraphs comes out still more
clearly when the table showing the displacements of the stations is examined.
In this, as in all the other lists, I have arranged the stations in order from south
to north, arranging those which are on nearly the same parallel in order from
west to east. It will be seen from this that the apparent displacements of the
three stations on the south of'the Khasi hills indicate an extension of 5 and
6 feet on either side of the starting point, Rangsanobo, Northwards of these, the
displacements decrease in a marked manner till we reach Mautherrichan and
Dinghei, to the north of which they increase conspicuously.

The true explanation of these facts I take to be, that the assumed value of the
base Rangsanobo—Taramun Tila was in excess of the real value. This led to an
apparent increase in all other sides which, so far as the calculations go, would be
cumulative and lead to an apparent expansion of the whole chain of triangles.
Northwards of Rangsanobo, however, this apparent expansion was masked by
an actual compression of the country as a whole, a compression which was not,
however, sufficient to reverse the apparent effect of the change in the base line.
Northwards of Mautherrichan and Dinghei the compression appears to die out
and the cumulative effect of the error of the original base becomes conspicuous
once more. | i

Another point to notice is that a distinct increase in the compression is
suggested as we go southwards. Towards the northern limit of the triangulation
its effect is small. Between the latitudes of Mosingi and Mautherrichan it becomes
more noticeable, and still more so between Mosingi—Mun and the three southern
stations ; the compression here being in two cases sufficient to completely counteract.
the apparent expansion. Itis by no means impossible that the compression was
gréater still further south, and reached its maximum at the southern limit of the
Khasi hills, so that the assumed unchanged base may have really suffered . a
greater actual change than any of the other sides.

We have seen that a detailed examination oí the apparent expansion of these
hills is consistent with a real compression and a shortening of the initial base. It
is also indubitable that such compression is consistent, not only with what is known
generally of the causes of great earthquakes, but also and more particularly with
the conclusion come to, on independent grounds, as to the cause of this particular
earthquake, while a real extension of the hills would be very difficult toreconcile
with that explanation. Putting these two considerations together the conclusion
may be adopted thatthe Khasi hills were not extended but compressed in a
north and south direction, and that the apparent extension was due to a real, but
unnoticed, contraction of the assumed initial base.

( 366 )

APPENDIX G. 77078967

In considering the changes in the heights of the stations it must be remembered
that the same dependence cannot be placed upon them as on the positions, réfrac-
tion being much more liable to introduce errors into vertical angles than in hori-
zontal angles. The changes, it will be observed, are, as a rule, small, but so far
asthey go seem to indicate a slight general upheaval of the hills, Even this
must, however, be regarded as doubtful. The starting point was Rangsanobo,
which was assumed unchanged, and calculating from it, Taramun Tila was found
to haverisen 6 feet. Ashasbeen explained the inherent probability is more in
favour of this station having remained unaltered in height than Rangsanobo, and
if this was so, then, assuming the results can be depended on, the tabulated
heights of all other stations would have to be reduced by 6 feet and as a result the
height of the Khasi hills would remain practically unchanged, or even diminished.*
As already mentioned no great weight can be attached to these observations,
but it is evident that over the area covered by the triangulation there has been no
marked alterations in the general height of the hills.

There aretwo stations, however, which form marked exceptions to this general
rule; Mautherrichan, with a rise of 24 feet, and Landau Modo with a rise of 17
feet. These amounts may beregarded as in excess of the possible errors of ob-
servation, and it may be taken that these peaks have risen by noteworthy amounts,
even though the tabulated values may not be exact to a few feet.

If we look to the situation of these two peaks the explanation of the exceptional
local elevation becomes clear. Both of them are situated on the crest of conspicuous
fault scarps, of the type described in Chapter IX, and Landau Modo is close to two
of the pools formed by distortion of the surface, details of which will be found in
the same chapter.?

If we turn to the changes in the heights of the secondary stations we find the
same thing repeated. The only two that seem to have undergone any conspicuous
change in height arethe Mairang and Kollong rock stations, both situated near the
edge of fault scarps. With the exact position of Laitbli, which has risen و‎ ft.,
I am unacquainted, but from its position it also is probably on or near the edge of
one of these fault scarps.

These facts regarding the distribution of changes in the elevation of the bills
afford a strong confirmation of the description of the structure of the Khasi
and Garo hills given in Chapter IX, and of the explanation given of the nature
of the focus and the cause of the earthquake given in Chapter X, but were not
available in time for incorporation in those chapters.

Though the explanation given above of the nature and cause of the displace-
ments that were observed is that which must be regarded as most probable, it
must not be forgotten that a different explanation is conceivable. The known facts
of structural geology show that certain regions have been subject to direct eleva-
tion which, so far as the surface rocks are concerned, is not due tothe indirect
effects of compression. Besides mountains of compression, of which the Alps and

1 The effect of errors in the calculated length of the sides, due to an error in the assumed
length of the initial base, may be neglected.
2 See p. 156.

( 387. /

368 OLDHAM: GREAT EARTHQUAKE OF 1897.

Himalayas are the leading types, there are mountains of faulting, the ‘ Schollen-
gebirge’ of German, and ‘ block mountains’ of American, geologists. In these
the relief is directly due to movements along the lines of faulting by which the
rocks are traversed, and the principal crests and steepest slopes, though modified
by subsequent denudation, follow and are due to these lines of fault.

In the case of hills of this type, their elevation far from having been due to com-
pression, has been accompanied by an actual extension of the ground they cover,
The faults are vertical or normal and the horizontal distance between two points
on either side of the fault, is greater after, than it was before, the elevation. The
mechanism of the formation of these mountains of faulting is not very easy to
understand ; it may be that the faulting and stretching are confined to the upper
layers of rock and are indirect results of a compression, and consequent thickening
of the more deep seated layers, through the small depth at which the level of no
strain is placed makes this supposition difficult of acceptance. A more probable
explanation is that over certain areas the upper layers are protected from that
compression which is their general fate, and follow the movements of the rocks
below the level of no strain, which are being stretched. The tension thus set up in
the rocks above the level of no strain would cause themto be split by vertical or
highly inclined fractures, and the horizontal extension would be taken up by sub-
sidence along these. .

Whatever may be the explanation, the facts are undoubted, and it is conceiv-
able, though with difficulty, that in such a region tensional strains might be set up
of sufficient greatness to give rise to a severe earthquake by their sudden relief.
The Assam range presents the superficial features of a region of mountains of the
character just considered, and it may be that the explanation of the fault scarps
is not that offered in Chapter X. Possibly the elevation of the range was accom-
panied by extension and not by compression, and in that case the apparent ex-
tension, exhibited by the results of the retriangulation, would be in accordance with
what should be expected. The small changes, exhibited by the southern triangles
would be explained by the absence cf fault scarps in that region, while the great
changes of horizontal distance, as well as of elevation, in the northern triangles
would be explained by the number of fault scarps seen in the region they cover.

If this explanation is adopted it would cause no change in the conclusions
stated in Chapter X as to the position, the complexity or the extent ofthe seismic:
focus regarded as a whole, but the assumed thrust plane or planes, uniting the
complex of more superficial fractures and faults, would not exist, and the strain to
whose relief the earthquake was due would be one of tension and not compres-
sion. In other words it would cause no change in what can be directly determined
from observed facts, though some change may be necessary in the more remote
deductions as to the ultimate cause of the earthquake.

! It would also make the faults, to which the fault scarps of the Assam range are due,
normal and not reversed faults, and the great monoclinal fold along their southern edge a
flexure, in the restricted sense in which the word is employed by Suess, and nota fold. The
distinction being t hat a fold is the result of compression, a flexure of differential elevation on
either side of the disturbed tract. As a consequence the distance between two points in any
given bed, situated on either side of the fold or flexure, will be less than the original distance
in the case of a fold, greater in that of a flexure..

(. ۹

APPENDIX G. 369:

. Though the possibility of a different explanation to that adopted in the text,
before the receipt of the results of the retriangulation, has been indicated, it must be
pointed out that though possible it is, so far as present information goes, much the
less probable of the two. In the first place, even if it be granted that tensional
strains might be set up in the earth's crust of sufficient magnitude to give rise to à
severe earthquake by their sudden relief, it is. almost impossible to understand
how their relief could give rise to such an extensive and complicated series of
fractures as would have to beinferred in the case of the 1897 earthquake. Then
the rapid variations in throw of the great Chedrang fault are more in accordance with
what would be anticipated if it had been accompanied, or caused, by compression
in a north and south direction, than if it had been the result of direct elevation.
The nature too of the observed displacements of the trigonometrical survey
stations bears out this supposition, that is to say the smallness of the apparent
extension of the sides which run more or less north and south, as compared with
those that run more or less east and west, is, as has been shown, explicable on the
supposition of a general compression of this tract in a N—S direction. If there had
been a real extension this should have been mainly north and south, seeing that the:
general run of the fault scarps is about east and west, and that the stretching of
the tract, if there is any, would be at right angles to this or more or less north
and south. The shape of the epifocal area points to the same conclusion, its length
from E to W being greater than from N to S, it follows that the displacements to
which the earthquake was due would be greater, on the whole, from north to south
than from east to west, and this whether they were the result of compression or
extension. The apparent displacements of the trigonometrical stations, if real and
due tc a general expansion of the hills, would necessitate this expansion being
greater from east to west than from north to south ; consequently this hypothesis is
not in accord with either the general structure, or the form of the focal area ;.
it has already been shown that both are in accordance with the alternative
hypothesis of a general, though not uniform, compression in a more or less north
and south direction.

It will be seen from this that the explanation of the cause of the earthquake
given in the text remains the most probable one, but that there is the possibility
of another explanation. The final establishment of one or other hypothesis will
depend on the extension of a retriangulation of these hills from a base known to
have been unaffected by the earthquake.

The triangulation carried out during tbe season of 1897-98 has fulfilled all that
could be expected of it. It has shown beyond doubt that there have been consider-
able displacements, both vertical and horizontal, of the hills within the epifocal
area, but as it could not be founded on any origin which was unchanged by the
earthquake, it can give no final measurement of the amount of these changes.

The facts established by the triangulation already carried out and the uncer-
tainty as to the exact nature of the changes that have taken place make it more
desirable that the work should be completed, and in ‘view of the possibility of
this it may be well to point out what is desirable.

DES c i [^ 3894)

370 OLDHAM: GREAT EARTHQUAKE OF 1897.

It so happens that the chains of triangles of the great triangulation are
singularly favourably situated for a measurement of the displacements that have
taken place. At the western end of the epifocal area a chain of triangles along the
go” meridian skirts the edge of the Garo hills and, so far as they extend, is based
on hill stations. On the g2° meridian another chain of triangles, that which
was partly reobserved in 1897-98, runs right across the Khasi hills. At right
angles to these two, a chain of triangles runs up the Brahmaputra valley, and in the
stretch between the two meridional chains, the stations are almost all situated on
rocky hills rising out of the alluvium.

A retriangulation should start on the 92° meridian, inthe country south of
the Sylhet valley. Any triangle here, whose three angles are all based on,hill
stations, would probably be found to be undisturbed, and might be taken as a
starting point. Should it be found to have suffered distortion, which is highly im-
probable, the observations would have to be carried backwards, away from the
Khasi hills, till an undistorted triangle was found, any of whose sides might
safely be taken as an initial base,

Starting from this, triangulation should be carried northward, till the transverse
chain of triangles inthe Brahmapuira valley was met. Here the work would
turn westward and continue down the Brahmaputra valley to Dhubri, and in this
part the greatest displacements are to be looked for. From Dhubri the triangles
of the series on the go” meridian should be reobserved as far south as they are
based on hill stations.

To complete the work it would be well to carry a chain of triangles from the
neighbourhood of Tura to that of Cherrapunji, thus linking the two meridional
series and forming a quadrilateral within the epifocal area. These triangles
should be based on peaks fixed by the triangulation of the topographical survey, a
triangulation which falls far short of that of the great trigonometrical survey in
rigour and accuracy, but being connected with it on all sides the maximum error
would fall far short of the displacements that are to be looked for in this tract.

The only previous occasion on which displacements connected with an earth-
quake have been actually measured is, so far as I know, that of the earthquake
of 17th May 1892 in Sumatra.! The principal triangulation was then in progress
and after the earthquake it was found that the angles observed would not
correspond with these observed before the earthquake. Certain of the triangles
were reobserved and it was found that the stations had been displaced, the greatest
distance between those which had undergone measureable displacement being
about 53 kilometres (33 miles). Beyond this some other stations were found to
have been displaced, but as they had not been fully fixed before the earthquake
their displacements could not be measured.

The greatest measured displacement was 1'23 metres with a probable error of
‘28m.; we may consequently say that the maximum displacement could not have

1 J. J. A.Muller De Verplaatsing van einige Triangulati-Pilaren in de Residente Tapaneeli
(Sumatra) tangevolge van de aardbeving van 17th Mai 1892:: Verhandl.¡K. Akad v, Weten-
schappen te Amsterdam, 1 sect. III, No. 2(1895) abstracted in Peterman’s Mittheilungen

1895, pp. 97-98.

(739073

APPENDIX H. 371

exceeded sft. With about the same length of triangles we have in the case of the
1897 shock a maximum displacement calculated at 12ft. As has been explained,
there is some uncertainty about the basis of this calculation, and if the explanation
offered above is the true one, this figure requires modification, but whether the
maximum displacement would be in excess or defect of it, cannot be decided in the
absence of detailed calculations. In any case it seems certain that the amount
must have been considerably in excess of 5ft., and it must be further borne in
mind that this triangulation lies near the limit of the epifocal area, and far removed
from the part where displacements were at their maximum.

Appendix H. a

A Review of the extent of the Lisbon Earthquake of ist November
1755, as compared with that of the Great Earthquake of 12th
Fune 1897.

In a preliminary note on the earthquake, written six weeks after the event, I
referred to it as‘ unsurpassed by any since the great Lisbon earthquake of rst
November 1755, and rivalling this in magnitude of the area over which it was felt;
surpassing it indeed, if we exclude the doubtful records of the earlier shock.’

This statement excited surprise and was even called to question in the press;
yet a fuller knowledge of the extent of the earthquake of 1897 has borne it out.
The surprise was a natural outcome of the very exaggerated ideas of the extent
of the Lisbon earthquake which are to be found repeated from one text book to
another without verification. I have thought it desirable, on this account, to
append a short review of what is known of the extent over which the Lisbon shock
was felt.

To begin with, no comparison of violence is possible. The data in the case of
the Lisbon shock are altogether too imperfect to permit of this, nor is it of import-
ance in determining the relative magnitude of the two shocks, Some earth-
quakes, it is known, may, like those of Ischia, have a great local violence but very
small extent, yet it would be absurd to speak of these as of greater magnitude
than others which, though nowhere rising to the same degree of violence, are felt
over a kingdom or two.

The only strictly scientific comparison of two shocks would be a comparison
of the energy developed in each. This measurement is unfortunately still im-
possible, but there is good reason to suppose that is approximately proportional to
the area over which they are felt. Any way this is the only possible means of
comparison in the present state of seismology, and as the results it leads to are
at any rate approximately true, it is the one that I have adopted.

Before instituting a comparison it is} necessary to have a clear conception of
what is meant by the area over which the shock was felt. The Lisbon earth-
quake originated under the sea and set up large sea-waves, to which much of the
loss of life in Lisbon was due, and these waves were observed everywhere along
the Atlantic coasts, even to the Weat Indies. Yet it would be altogether mislead-

i | | | (29m)

372 OLDHAM: GREAT EARTHQUAKE OF 1897.

ing to include, for purposes of comparison, the whole of the Atlantic in the area
over which the shock was felt. The mode of initiation of these great sea-waves
is not yet fully understood; they evidently are only produced when there is a
sudden displacement of a large mass of material on the bottom of the sea, and
the most probable hypothesis is that this displacement is in the form of a sub-
marine landslip. It is certain that there is no direct connection between the
magnitude of the earthquake and the size of the sea-wave. Great earthquakes,
even when they originate beneath the sea, may give rise to quite small sea-waves,
on the other hand, large and destructive sea-waves may result from quite mode:
rate earthquakes. In either case the wave once set up will travel till it reaches
land, and the area over which it may be observed is only limited by the size of the
ecean.

Rejecting the sea-wave as useless for purposes of comparison, we may turn to the
consideration of the area over which the shock was felt on dry land. Throughout
Spain and Portugal it was felt with varying degrees of severity. In southern
France the shock was distinctly felt and appears to have been noticeable even in
Normandy, but it does net seem to have been felt anywhere in England.

This statement may be considered extraordinary in face of the often repeated
statement that the Lisbon earthquake was distincly felt in the Derbyshire mines.
As this appears to be one of the accepted facts of seismology, its contradiction
requires support, and I consequently reprint here, verbatim, the original authority
on which the statement is based.

í

LIX.—An Account of the Earthquake, 1st November 1755, as felt in the Lead
Mines in Derbyshire, ina Letterfrom the Reverend Mr. Bullock to Lewis
Crusius, D.D., F.R ہک‎

ASHFORD, NEAR BAKEWELL.

Derbyshire, grd March 1756.
Rev. SIR,

I here send you an account of the earthquake which happened at the lead
mines on Eyam-edge inthe peak of Derbyshire, on Saturday the rst of November
1755, about 11 o'clock in the forenoon. I made a strict inquiry at the mines, on.
the 21st of February last, both for my own and your satisfaction, and can assure
you that the circumstances related may safely be relied upon as matter of fact. -If
there be anything in my power which you would choose to have explained more
particularly, please to communicate, and it shall be done with the greatest pleasure
TR E

SIR,

Your most obedient and obliged humble servant,
(Sd. WILLIAM Buttock.

Francis Mason, overseer, says, that he satin alittle room, which he uses to
write in; it stands about forty yards from the mouth of one of the engine shafts,
He felt one shock, which very sensibly raised him up in the chair, and caused
several pieces of lime or plaster to drop from the sides of the room; the roof of it

(252. 3)

APPENDIX G. 373

was so violently shook, that he imagined nothing less than the engine shaft was
run in; whereupon he immediately went out to see, and, contrary to his expecta-
tion, found the shaft open, and all things about the spot in their proper order.

Upon inquiry, I was informed by the aforesaid Francis Mason, that in a field
about 300 yards from the mines, there had happened achasm or cleft on the
surface of the earth, which was supposed to be made at the same time he felt the
shock, for the following reason. It lies close by a road, which he uses daily to
and from the mines: in the morning when he came, there was nothing uncommon
to be seen, but on his return at evening he observed a cleft about one foot deep
and six inches in diameter; its cOntinuation from oneend tothe other was near
150 yards, being parallel to the range of the vein on the north side. When I
examined it, which was upwards of three months after the shock happened, the
length of it was not much more than 6o yards, though I could perceive some
vestiges of its further continuance : the depth of it was about eight or nine inches
and its diameter four. Asthe soil was very light, and the season has been re-
markably wet, it is highly probable, that the fissure is considerably closed since it
was first made. These were the most remarkable circumstances which happened
upon the surfaceof the earth. Though my enquiry was of every one in particular,
that was there employed about the mine, the concurrence of whose testimonies
might seem more strongly to confirm the account, yet I look upon it as unneces-
sary to trouble you with every man’s story, which would be only a repetition of,
or something similar to, what has been before related.

William Hallom, and Jo. Howson, miners, say, that at the aforesaid time they
were employed in carting or drawing along the drifts the ore and other minerals
to be raised up the shafts. The drift, wherein they were working, is about 60
fathoms, or 120 yards deep, and the space of it from one end tothe other 50 yards
or upwards. Hallom was at the end of the drift, had just loaded his cart, and
was drawing it along, but was suddenly surprised by a shock, which so terrified
him, that he immediately quitted his employment, and ran to the west end of the
drift to his partner, who was not less terrified than himself. They durst not
attempt to climb the shaft, lest that should be running in upon them, but consulted
what means to take for their safety. Whilst they were thinking of some place of
refuge, they were alarmed by a shock much more violent than the former; which
put them in such a consternation, that they both ran precipitately to the other end
of the drift. There was a miner working at the forfield, or east end of the vein
about six fathoms below their level, who called out to them, imagining they were in
danger of being killed by the shafts running in upon them which he supposed was
the case; and told them if by any means they could get down the shaft to him,
they would be more secure, because the cavity where he was working was encom-
passed with solid rock. They went down the shaft to him, where, after observing
they had neither of them received any. misfortune, he told them, that the violence
of the second shock was so great, that it caused the rocks to grind one upon
another. His narration was interrupted by a third shock; and after an interval
of about four or five minutes, was succeeded by a fourth; and about the same
space of time after, by a fifth; none of which were so violent as the second.
They heard after every shock aloud rumbling in the bowels of the earth, which

ری ورڈ کا

374 OLDHAM: GREAT EARTHQUAKE OF 1897.

continued for about half a minute, gradually decreasing, or appearing at a gteatef
distance. They imagined, that the whole space of time, from the first shock to
the last, was about twenty minutes; and they tarried about ten minutes in the
mine after the last shock } when they thought it advisable to examine the passage
and to get out of the mine, if possible. As they went along the drifts, they
observed that several pieces of minerals were dropped from the sides and roof, but
all the shafts remained entire without the least discomposure.

The space of ground at the aforesaid mines wherein it was felt, was 960 yu ds
which was all that was at that time in workmariship.!

From this account it is evident that we are dealing, not with the Lisbon earth-
quake, but with an independent one of local origin. Too much importance must
not be attached to the reported time, which is an hour later than that of the Lisbon
shock, as it is evidently only approximate, but the shock described is a fairly severe
one, much severer than indicated by any of the accounts of the shock in or out of
England, till we get to the south-western part of France.

Besides the severity of the shock there is the fact that it was"felt underground
almost as severely as at the surface. With regard to earthquakes generally
it seems well established that, in the area of propagation, the motion that can be
felt is confined to the surface and, at a very small depth below ground, it ceases to
be felt. Within the area of the epicentre, however, or even within that of the earth:
quake core, there is no reason why this should be so, and the account of what was
felt in the Eyam-edge mine points to its being on the direct path upwards from
the focus to the surface; a supposition which is borne out by the large vertical
component of the motior indicated in the overseer's account. |

The Derbyshire earthquake may consequently be excluded from the accounts of
the Lisbon earthquake; at the miost it can only be connected with the greater
shock as a sympathetic earthquake, and even that is doubtful.

With the exception of the Derbyshire earthquake all the other accounts col-
lected by the Royal Society, and published in the 49th volume of its Philosophical
Transactions, relate merely to the effects of the sea-waves aleng tbe coasts, or
oscillations of ponds, canals, and even lakes, similar to those described in Burma?
in the case of the 1897 shock. One account alone mentions the earthquake as
having been felt near Reading; Mr. Paunceforth's gardener. stated that he felt |
a most violent trembling of the earth for fifty seconds. All the other accounts
either definitely state that no shock could be felt, or, in a few instances, say
nothing as to this, and it is probable that the violent trembling felt by Mr,
Paunceforth's gardener, who was watching the washing back and forwards of
water in a pond, was the result of imagination.

In the accounts of the shock as experienced in Holland and Denmark,I can
find nothing to support the idea that it was felt there, only oscillations of water in
ponds and canals are recorded ; while the time at which the shock is said to have
been felt in southern Norway, 4 A.-M., shows that the shock, which was certainly

1 Phil. Trans. XLIX, pp., 399 ff.
2 See chapter Il, page 39.

MË,

APPENDIX G. 375

felt there, was a distinct one. The reputed feeling of the shock in Iceland and
Greenland, which has always been regarded with doubt, may safely be rejected.

From this it seems that the most northerly place at which the shock is reported
to have been felt was Reading, and even this is so doubtful that it may be rejected.

In an easterly direction it is said to have been felt in parts of Switzerland, and
though the accounts refer mainly to disturbances of the water of the lakes, there
Seems good reason to suppose that the shock was sensible in parts of western
Switzerland. In Italy it is said to have been felt at Milan and 8 leagues N. N. W.
of Turin, but it is not clear whether it was actually felt, or only recognised by
its effect in setting hanging lamps aswing. It must be remembered too that the
Italian, besides being gifted with a lively imagination, is, by descent, a skilled
observer of earthquakes. Even our earthquake of 1897 is said to have been felt
in Italy." In any case Milan and Turin can hardly be looked on as part of the
main area over which the shock was felt, but as an outlying area like that round
Ahmadnagar in the case of the 1897 earthquake.

To the south, it is impossible to define the area over which the shock was felt
with any degree of accuracy. It was very severely felt in the north-west corner of
Africa. As far south as Morocco a great number of houses are said to have fallen,
and one of the hills near by to have been rent in two. This account is almost
certainly greatly exaggerated, for at Funchal, in Madeira, the shock was only
smart enough to make windows rattle.

To the west we have the Atlantic ocean, and it is impossible to say how far in
this direction the shock might or might not have been felt had there been dry land.

The position of the epicentre cannot be fixed with the same accuracy as in the
case of the 1897 shock, but the centre from which the great sea-waves started was
fixed by David Milne at about 39° N. Lat, 11° W. Long.? The epicentre of the
earthquake was not necessarily the same as the centre from which the sea-waves
started, but the two are not likely to have been widely separated, and there will be
no material error in assuming the two identical.

Taking the epicentre as situated in the position indicated, we find that Milan is
distant about 1200 miles, or practically the distance of Ahmadnagar from the
epicentre of the 1897 shock ; and Reading about 1000 miles, against a distance of
over goo from the epicentre to Bezwada, where the 1897 shock was distinctly felt.

From this it would seem that the distances from the epicentre at-which the two
shocks were felt were practically the same, but as I have explained above, the
accounts from Reading and Milan, especially the former, do not appear to be

1 At Leghorn a tremor is said to have been felt by some, though not living in upper stories,
at about 12-15. At Catania some people are said to have noticed it, but the accounts are
vague and uncertain. At Spinea di Mestre (near Venice) an undulating shock of 4 seconds,
is said to have been felt in the Observatory at 12 h. 18 m. 39 s. (Boll. Soc, Sismol. ltal., 111, pt.
ii, pp. 251, 285, 289.) The times are mid European, and exactly one hour fast of Greenwich
mean time. They correspond with that of the first phase, and it is possible that these waves,
of half-second period, were just sensible to very skilled observers, favourably situated.

2 Edin, new Phil. Journ., XXXI, 263 (1841).

( 355 5)

27 OLDHAM: GREAT EARTHQUAKE OF ۰

such as can be accepted for comparison. If we exclude them, the area over which
the shock could have been felt is very much reduced ; a rough estimate which
I have made, and nothing more is possible, places the total area over which the
shock would have been felt, had it all been dry land, as not more than 1,000,000 `
square miles. A similar estimate for the 1897 shock gives about 1,750,000 square
miles.

These estimates cannot be regarded as giving an exact ratio of the magnitude
of the two shocks. Inthe case of the Lisbon earthquake the estimate must ne-
cessarily be very approximate, but it is not probable that it could be raised above
that of the earthquake of 1897. At any rate the facts fully bear out the statement
that the shock of 1897 rivalled that of 1755 in magnitude, and if the doubtful
records of the latter are excluded, its magnitude was certainly greater.

There is one feature of the 1755 records which might lead to a different idea
and that is the many accounts of its effect on ponds and lakes even as far north as
Scotland and Denmark," as contrasted with the absence of similar accounts in the
case of the 1897 shock. This is, however, easy of explanation. In a westerly
"direction the country which would occupy the position of England relative to the
area over which the shock was sensible, is mostly desert, sparsely inhabited and `
devoid of sheets of water. To the south-westwards, in the Peninsula, artificial
ponds and lakes are not wanting, but there is an absence of educated and intelli-
gent observers. To the east and north the area over which the shock was felt
extends into country from which no information can be obtained.

On the other hand, when we consider that the 1755 shock extended over a
populous and civilised country, that it attracted. universal attention, and that all
the leading scientific societies and academies of western Europe set themselves to
collect information regarding it, and when we consider how comparatively few and
imperfect were the observations collected, even as regards this one point of the
disturbance of water in ponds, it is not surprising that in the case of the 1897
shock they are altogether wanting.

That similar oscillations of water in tanks took place I do not doubt; we have
accounts of them within the area over which the shock was felt, and besides those
mentioned in the text I received numerous accounts of disturbances of water in
wells having been noticed. "There are, besides, two separate accounts, showing
that, outside the area over which the shock was felt, the undulation of the ground
was sufficient to make doors swing to and fro.?

For these reasons it does not seem to me that the accounts of the effects of the
Lisbon earthquake point to its being of greater magnitude, so far as this can be
measured by the area over which the shock was felt, though it is possible that the
wave motion was of somewhat different character and gave rise to surface undu-
lations of greater size and steepness.

1 The reputed disturbance of the North American lakes is too doubtfully connected with
earthquake to be considered.
2 Chap. II, p. 37.

( 57€ )

APPENDIX I, 377

Appendix I,
Bibliography of the Earthquake.

So great an earthquake has naturally given rise to a large volume of literature
not only in the shape of notices in the daily newspapers, but also in scientific
magazines. Nearly all of these are, however, written at second hand, and, so far
as I know, the following list comprises all that can be quoted as original author
ities :—

1, Thos. Heath; An Edinburgh Record of the Indian Earthquake: Nature,
LVI, 174 (1897)- -

A letter to Vature, giving a brief account, and an illustration on a reduced scale,
of therecord of the earthquake by the bifilar pendulum at Edinburgh. A letter
to the same effect was published in'the Calcutta Englishman of 6th July 1897.
Both may be regarded as superseded by No. 8.

2. T. D. LaTouche. The Calcutta Earthquake; Nature, LVI, 273-274.

This is a brief account of the earthquake and its effects in Calcutta, written
yery shortly after the occurrence,

3. The Calcutta Earthquake: Nature, LVI, 346 (1897),

A short notice, and reproduction of part of the record of the horizontal pendu-
lum instrument at the Royal Geodynamic Observatory of the Rocca di Papa.

4. T. D. LaTouche: The late Earthquake in India: Nature, LVI, 444-44
(1897).

Anextract from a private letter describing the effects of the earthquake in
Assam. It expresses the idea, natural at first, but which had afterwards to be
abandoned, that the origin of the earthquake was connected with the great uni»
clina] fold or flexure which bounds the southern edge of the Assam hills.

5. J. Milne: The Recent Australian and Indian earthquakes; Symons’
Monthly Meteorological Magazine, XXXII, 92-93, 1897.

Contains a statement of the times at which the effects of the earthquake were
registered at Edinburgh, Shide I, W., Strassburg and Grenoble.

6. T. Moureaux : Bulletin du Bureau central ; quoted in Cosmos, VI, 652
(1897).

Records an interruption of the traces of the magnetographs in the observatory
of Parc St. Maur. I have not been able to refer to the original.

7. R. D. Oldham: The Earthquake of 12th June: Records of the Geological
Survey of India, XXX, 132-124 (1897).

A brief preliminary notice, written about a month after the earthquake, giving
some of the more striking facts concerning it,

8. M. Baratta : Il grande terremoto Indiano del 12 giugno 1897: Bolletino
della Societa geografica Italiana, 3rd series, X, 1—8, and 1 plate (1897)

An account and illustration of the records of the earthquake on the micro-
seismograpks of the University of Padua, designed and controlled by Prof, G,
Vicentin*,

( 377 )

378 OLDHAM: GREAT EARTHQUAKE OF 1897:

9. Thomas Heath: Note on the Calcutta Earthquake (June 12th, 1897) a$
recorded by the Bifilar pendulum at the Edinburgh Royal Observatory: Pro-
ceedings of the Royal Society of Edinburgh, 1897, 481—488, 2 Plates.

An account and reproduction of the record of the earthquake on thg instrument
referred to in the title.

10. N. A. F. Moos: A short Note on the disturbance of the Magnetical and
Meteorological Instruments at the Colaba Observatory during the Earthquake of
rath June 1897: Magnetical and Meteorological observations made at the Govern-
ment Observatory, Bombay, 1896 ; Appendix A., pp. (1]—[7]. Bombay, 1897.

gé paper is fully noticed in Chapter XI.

. F. Omori: A preliminary note on the Indian Earthquake of June 12th,
Hor Indian Daily News, 23rd October 1897.

A newspaper article by Prof. Omori, written after his tour through Assam and
Bengal. He estimates the depth of the focus at 20 miles, an estimate apparently
based on the supposition that the focus was of comparatively small horizontal
extent, and on the assümed small number of aftershocks. As has been shown
the nature and form of the focus was so different from that of the Japanese Earth:
quake of 1891, that the analogy does not hold good,

12. Report onthe Earthquake of 12th June 1897, so far as it affected the Pro-
vince of Assam. |

The official report of the Government of Assam, signed by Mr, E. A.” Gait;
Officiating Secretary to the Chief Commissioner of Assam, pp. garid 72, issued asa
separate report, reprinted from the Assam Gazette.

13. A. Cancani: I pendoli orizzontali del R. Osservatorio geodinamico di
Rocca di Papa, ed il terremoto Indiano del 12 giugno 1897: Bolletino de la
Società Sismologica Italiana, 111 pt. i, 235-240, and 1 Plate (1897).

A description of the records of the instruments referred to inthe title. The
Plate is a reproduction of the same diagram as is represented on Plate XLI of
this memoir. ;

14. Report on the Earthquake of 12th June 1897 in Bengal.

The official report of the Government of Bengal signed by Mr. C. W. Bolton;
Chief Secretary. Calcutta Gazette Supplement, 5th June 1898.

15. H. Luttman-Johhson : The Earthquake i in Assam; Fournal of the Society .
of Arts, XLVI, 473-493 (1898).

A non-scientific account of the earthquake, containing also a note by myself
on some of the scientific aspects of it. )

16. C. L. Griesbach : General Report on the work carried on by the Geolo-
gical Survey of India for the period from 1st January 1897 to the 1st April 1898 :
8°, Calcutta 1898,

Contains at pp. 16—18 a note by myself practically identical with that included
in No. 15.

17. G. Agamennone: Il terremoto del India del 12 giugno 1897: Atti della
Reale Accademia dei Lincei : 5th series, VII, 265—271 (1898).

A discussion of the records of the earthquake on the self.recording . seisrrio*
graphs and magnetographs in European observatories,

( 375 )

APPENDIX I. 379

18. G. Agamennone: Il terremoto dell India del 12 giugno 1897: Bolletino
délla Società Sismologica Italiana, IV, Pt. i, 33— 40 (1898).

A short note, based principally on Nos. 1,2 and 6. `

19. G. Agamenncne: Notize sui Terremoti osservati in Italia durante l'anno
1897. (1° semestre.) Bolletino della Societé Sismologica Italiana, III, Pt. ii
(1867).

A list of all shocks recorded in Italy: pages 249 to 293 contain detailed,
accounts of the records of the Italian observatories, and more brief reference to
those of other European observatories. The results are discussed in—

20. G. Agamennone: Ecoin Europa del terremoto Indiano del r2 giugno
1897 : Bolletino della Societé Sismologica Italiana, 1۷, Pt. i, 41—67 (1893).

A discussion of the records of the Italian seismographs, which was received
too late to be utilised in the preparation of Chapter XV, but has been referred
to therein,

21. G. Grablovitz : Rettifica alla relazione dell’ oSservatorio geodinamico di
Casamicciola (Ischia) sul terremoto dell' India del 12 giugno 1897 : Bolletino
della Societe Sismologice Italiana, pp. 167—168.

22. G. Agamennone : Replica alla rettifica del Prof. G. Grablovitz alla relazione
sulle osservazioni fatte a Casamicciola in occasione del terremoto Indiano dell 12
giugno 1897: Bolletino della Societe Sismologica Italiana, pp. 169—172.

( 379 )

INDEX TO PLACE NAMES OF INDIAN LOCALITIES.

(Places of minor importance whose position is suffisiently indicated
by the context have not been indexed.)

288.

36, 44, 65, 71, 74, 198.
67.

66, 71.

118, 136.

20.

110, 203.

34, 192.

Long. E.
72 45
gr 23
go” 37
or Sa
E
92° 38’
gi? 16’
92" 56’
94* 17
89” 35
81% 55'
95” 17
"oH
go* 24^
go” 47’
ap 2
88° 18'
89 Gi
92 37
899 41’
88° 43’
86? 53

Lat. N.
24° 36’
23? 5o
e, Si
26° 48’
277. 10"
29... 4
23° 52
20 7
"TT,
26° 30°
ant 36
19 21’
"وو وھ‎
257 34’
24° 38
24° 8
24 14
25° 40’
24° 53”
20° 46”
264

a oa E A
Agartala ۰ E . .
Agia . e . e ` .
Ahatguri . A e S `
Agra . . e > e e

Ahmedabad E e o $
Akhaura : 7 e e 3
Akyab ^ « e e

Alguada . " D " .
Alipur, part of Calcutta : M
Alipur Duars

Allahabad P $ -

Allan myo . : a . e
Anand . S ٠ .

Arbela e . e . .
Atrabari : 5 H ^ e

Atrai . . . - . LJ

A zimganj D D . . e
Badarganj . H .
Badarpur. d و‎ A i
Badraghat . 5 3 t £

Bagula ` GB Ae, :

, Bakarganj district, capital Barisal .

Balasor A ,

EH

Page,

| 97*

84, 112, 117.
37°
325.
327.

| Ilo.

110, 313, 324

66, 71, 74.

200, 315.

50»

288.

65, 71.

311.

35, 67.

43, 109, 302, 325.

40, 193.

193. =

IIO, 327.

127, 169, 261.

67, 71.

99, 107, 109, 321.

19.

37, 57, 61,70, 180, 241.

124, 148, 334°

211.

80, 127, 161, 169, 196,
266, 334.

332.

60.

-—

Balipara E 1 3 4
Bálpakrám . e ‘ E
Bangalore. y + š £
Banka . MG A Dadder,
Bankipur œ e : :
Bankura ` E : d

Bardwan E d 3 $

Bareilly ` : E E
Barisal , E EL. d
Baroda . e او ا‎ PA, :
Baxa e 3 y E $
Benares ا‎ a 0:
Berhampur . ` A y
Bezwada > : ME

Bhagalpur . x E 3
Bhamo : me che A

Bhartpur œ A d =

Bihar . 2 3 E 2
Bime Klee ا سیا‎ ie
Bilimora rae d 3

Birbhum district, capital Suri
Bogra . e . . .
Bogribari œ e SE
Bombay ^ . e ^
Bordwar e . . e
Borjuli . . . e .
Borpeta » . e . e
Brahmanbaria . . .
Budge Budge . . e
Burhanpur . ° . .

Buxar . e a a S

INDEX. 111

Lat. N. Long. E. Page.
Cachar district, capital Silchar e . 2 ا‎ 109, 110, 343.

30, 43, 57» و‎ 192, 231,
257, 277, 2:9, 315, 4۰
Cawnpore . è - . . Ac 265 287 So? 24^ | 65, 71 pose
Chaurha Hat 4 - - e : 2060 8 89" 32’ | 287.

Chandbali e wh Ne D D P 20° 46’ 86° 48’ 59.

Calcutta . ek e . . 225 34 88? 24

Chandpur . STE . . E o] oo" 42’ | 333.
Chapra . e . . e . 25° 47’ 84? 47’ 65, 71.
Cheran e . , o Po 28?" 45 Qo" 46" 1158:

Cherrapunji . . e . ۰ . 25° 17 91° 47° | 78, 82, 116, 209, 272, 318,

346, 356.
Chhutuk e = e . ^ a aem gt 91? 42' | 208, 318, 356.

EDS IL ONCE OC OW COURSE 92 3 |344.
Chutapong. V. ^0 7972, KE ORO A S ant 91? 53' | 30, 41, 64, 71, 110, 300,
Coronada OR: AT AC EA D y 82 13’ wi. `
Colaba (part of Bombay) ` . . - ... T

Colgong . : . . e nko "eil 87° 17’ | roo, 326.

Comilah . et S B ok. oam a 91? 13’ | 208, 345.

Dabhoda . . * pU v ada wë SCH EC, 72° 49’ | 66,71.

A al e St, Zoe "ap 90? 26’ |109, 290, 318, 329, 345.

Dalbot . ^ . : e e al 25". Ser go 44 141,

Dalton cabs] e i "én BE VIN کر‎ "ve, B gy is

Dalu . 4 e - e z SE 35" mn Qo" y T xm

Damra e » e e . 4| 25 54. go” 50” | 128, 147, 158, 197.
Damukdea . ٠ - a e vk کرو‎ s 80” 4 | Gay 9s

Darangiri , E $ * e 4 25° 8۶ 90? 58’ | 127.

GEI وت روصت‎ NOE IUS QUU "A NI eet X go” 40’ | 153.

cuoc DOCE E P EMI OPERE ET ge 88° 19 | 43, 63, 71, 119, 191, 211,
Darrang district, capital Tezpur . : 2 sae x go

Dauki . e ^ á z ý یا رہ‎ ee $27. 3^ "1-129.

Debra ۰ e e " D ۰ . 22? 24' 87? 75 325.
Dehegam e D $ D . . 237 ri 72? 51 67, 71.
Delhi . - : ` > = ودنہ‎ 77 IG NOG Th.

————
-a سے‎ M ——

1v INDEX.

e EE
Lat. N. Long. E. Page.
Deogarh ` e e e 24° و‎ 86° 45° | 110, 327.
Bhana A. A oe ES GE d 91° 59’ | 344
Dhubri . ۰ e . ` 260 ai o 2’ | 18, 78, 102, 127, 163,
S 211, 259, $17, 33»
; S j 345, ۰
Diamond Island . . > ES ty 94 19’ | 51.
Dibrugarh + : : : 27° 28 94° 57' | 163.
Dilma . : . e : 25? eat 90? 43’ | 80, 142.
Dinajpur E e e e کو‎ 357 88° 41’ | 109.
Disang Mukh ۰ e e apa S 94° 35’ | 163.
Diwangiri à . : . 26° 51’ 9i? 27’ | 335.
Dobu . ` e e . 25° 34 9o? 46’ | 155.
Dobukhel . ۰ . e 25° 29' 9o? 47’ | 152.
Dowke, see Dauki © Lé e eee zent e...
Dumka e s a] c UR 24^ 16 87° 17 | 327.
Faizabad + - e . 26° 47' 822 12’ | 65, 71.
Faridpur e . E . 2 36 89° 53’ | 110, 317.
Gaibandha . a . 25 20' 89 37 | 319.
Garo Hills district, capital Tura A I 8, 80, 84, 92, 103, 112,
115, ۰
Gauhati : ` . > 260 1 91? 48’ | 78,81, 100, 107, 113,
126, 133, 164, 178. 199,
209, 263, 317, 325.
345, 355.
Gauripur . e . e 26° 11 9o? y” | 19, 261.
Gaya . . S : A 24° 49' 35 1031-64, 715 193.
Ghaziabad + e e : 28? 40’ 77? 28' (66, 71.
Gheor . ٠ e . D 23° 54’ 89° 56’ `
Giridih . e e ` E 247 10" 86° 17’ | 36, 64, 71.
Gitaldaha . . 2 e 269. 2 8G, 321 | 280.
Goalpara district, capital Dhubri aes zt
Goalpara e ? a 20" 51^ 002. 41^.| 19, 43, 75; 78, 81, 103,
107, 109, IIO, 127,
133, 160, 261, 317,
333, 345: 355.
Goalundo . e a ^ 25^ 50 89? 48' 3. ds :
Goas . ; $ ; > 24° Io’ 88 39 | 428.

INDEX. y

Page.
Gobardangals U TT 315, 355.
Godda. .. . . > ۰ 110.
Golaghat e . . . . 341.
Gomastapur . 5 : e 327.
Govindganj . |. + > ۰ : 319.

Gwalior e a” . e D 66 71, 110,

e 343.‏ 4 00004 ای
Hajo, see Hazu . " ٠‏

Ek, PESA

103, 172, 206, 280,

Harirampur , . ۰ » ` 331.

Hathimora . di 162.

Haza . . " ۰ e e 267, 334.
Helem, e ا‎ ARTIR 339.

Henzada . e . . e 67.

Hooghly, see Hugli m
Hoshangabad ra e 606۰

Howrah - : 2 e - 110, 290, 325.
Hugli e . è ۰ e 63, 71, 110, 325»

Hukong Valley . ۰ e ۰ SI,

Isvarganj . D D ۰ . 20, 102, 331,

Jabalpur e . . . e 193.
Jaidebpur . e ۰ . . 330.
09+8 171,
Jaipur ۰ " A 66, 71.
Jaipur as e 206.
Jalpaiguri . ` ` e - 281, 354.
Jamalpur . ۰ . e 331.
Jamalpur . ^ . ` ۰ 305.
Jangipur . i ` : e 329.
Jatrapur ہ.<‎ e . » ٠ 259-

vi INDEX.
Lat. N. Long, E. Page.
Jellinghi . ` . e e 243 8 88° 46’ | 328.
Jessor e ` ` è e 23? 10' 89? 15' | 110, 315, 328.
]henidah . e : e 202 89 14 | 328.
Jhira . ` e . : ۹ 25201557 go” 41” | 143, 1406.
Jorhat e > . . - 26° 46’ 94° 16° | 340.
Kalewa . x ` ` . 23 CH 94° 18 | 67.
Kalna ` e . - . 239 13 88° 24 | 324.
aseene uo wb ا‎ pes 26° ou 91° 45 | 334. |
Kamrup district, capital Gauhati a Le he 27, 100, 139, 160, 3340
Kandi A ` 2 e ` 23? 58 88? 5’ | 329. SN ۱٠
Kanrut = یی‎ = à SE 25 91? 34 157.
Kapasia Mere eh کت ںاہ‎ 24 ur. 90? 8 | 329.
Karimganj . e " e ` 25) 52! 92^ 24' | 343.
Kasva : - e $ a 23° 45’ 91? 13’ | 333.
Katha 7 i 1 s E 249. 10 | 06 1237 ose
A A 2 7 24° 53 92° 38’ | 343.
Katmandu. e - . . 27° 42 85 12 28.
Katwa = - ` : e "y 39 88° 11 | 324.
Kaunia e ` - š ~ 25% 47’ 89° 20’ | 107, 126, 172, 286.
Keraniganj . ` E و‎ > 23° 43 90? 24” | 329.
Khana e 2 Np ees e 23° 19 87° 50 | 64, 71.
Khandwa . . . : . 212 So 76-23" 37.
Kharipput <i 2^ p d.d E 22° 21 87 24' | 67.
Khasi Hills district, capital Shillong 7 BER 85, 92, 112, 115, 118, 120,
Khulna - : e e ^ 222 وہ‎ 89 37: E
Kindat ا را‎ AS ee 98". -| 57.
Kishorganj . ° e a e 24° 26 90? 40’ | 331.
Kistna district, capital Masulipatam e ... 110.
Kohima ee eee Ah 25 4 | 94 10 | 27, 119, 194, 342.
Kotah e 2 e 4 s 35° 10 255. 52! 66.

————

Page.
Kotgarh e . e . . 193.
Krishnagar . . . . . 43, 207.
Kuch Bihar lo 9 2 e 25, 43» 75, QI, 126, 127,
Kurigram . à " « e ord vix
Kyauk Pyu . . R ۰ . 41.
Lailangkot ( Laitly nkot) A e 116.
Lakhimpur . e . e e 110, 340.
Lakhipur ` ž . e . e e 127.
Lalitpur e . . . . 51
Landomodo e ? . . 156.
Laubersat . . . . . 117
Le myethna e . . . 67.
Lenkra ^ e " e e 154.
Lhassa . . e ` ۰ 38.
Lobpur ۰ . . . ۰ 325.
Luckeeserai + e e = e 64, 71.
Lucknow - 7 $ = . 36, 65, 7t.
Lumding e e e . e 194.
Madhipura . . . e . 325.
Madhupur jungle . . e e 109, 292.
Magura e . . . e 328.
Mahiganj . e . ٠ . 320.
Maimansingh e : e . 20, oi, 107, 109, 126,
135, 292, 331, 355-
Mairang . . . . . 157.

Maldah D e D D D 435 109, 327.

Mandalanggiri . ° e . I5lo.
Mandalay . . e . e 193.
Mangaldan e ye te es 26, 339»
Manikganj . . e . ۰ 329.

Manipur E e . e . 28, 110, 118, 343.

viil

Page.

I5, 161, 261.

135, 157, 371.
273.

117, 136.

195.

328.

33, 110, 192, 325»
119»

65, 71.

319۰

325۰
21, 294.

329.

43, 110, 311, 328.
52.

51.

320.

40.

64, 71.

65, 71.

27, 102, 334.
$13.

nc a a —————————————————

Long. E.
89° ss
91? Sol
91° 48
BI uM
nt 3r
96° 13
88? 0ھ‎
83^. 21"
93 o
82? 38'
89? 20'
89 31
9I? CC
go” 59
86? 30
go” 4o'
87^ 49
go” 19'
go” 35
88? TO
96° A
96° 27
88? 25
DES FT
89° 46
96° 14
88” 28
81? 56’
91^ 30’
87° 54

INDEX.
Lat, N.
255. 52"
25° 27'
257.15
MO II
25° 32
24^ 281
23 47
22? 25"
26? 15'
25° 16"
25° 34
26 0
23° 50’
25° 32
25% 23
237 47
23^ 59
24° 46
23 33
z4 a
20° 45’
25° 24
28 25
23:03
25” 58
24^. 26
A
25° 54
26 37
24° 19

capital Kohima

۰

Mankachar . E
Maophlang .
Maosmai

Masulipatam e

Mauterichan (Maomadraishan)

.

Mawlu e ‘
Meherpur ` e
Midnapur + .
Mikir Hills . `

Mirzapur

Mithapukar . e

Moghal Hat e 6

Mogra e .
Mokersa e .
Monghyr . e
Monohordi . .
Moweswar . e

Muktagachha :
Munshiganj .

Murshidabad -
Myelat St,

M yitkhina e -e
Nadiya e e
Nabinagar . e

Naga Hills district,
Nageswari

Nahakaung . A

Naihati . e
Naini , e .
Nalbari . .
Nalhati e e

INDEX. ix
Lat. N. Long. E. | Page.

Naphak ^ 2 es X A 2 0 a5 Qo? 43’ | 155, 196.
Narainpur +». + . e > 26° 55’ 93° 55’ | 340.
Narsinghdi » » . = E 1 29° 56 Qo? 47’ | 329.
e i.a t» dub. 2X. او‎ 09 0 | 332.
Nawabganj e e : ° 23° 40’ 9o? 13/ | 329.
Nepal, capital Katmandu. eil: A A 38, 10g.
Netrakona » > SE ee A. کو‎ Se 9o? 48’ | 331.
Nilphamari . á - e A MA at En 88° 54’ | 98, 172, 280, 319»
Nimach e e - A A .] 24° a8 74 SA | So.
Noakhali . ap d . » elt agi ane 91? of | 336.
Nongpoh à ۰ e . e alt) 3:20 gat 91? 56’ | tgo.
Nongstoin . - e e - - 9527 7و‎ 91? 20’ | 130, 136.
Nowgong + - e ^ : 4|. 28". ax? 92? 45’ | 103, 109.
Pabna ۰ : e . ; datis 89? 17” | 109, 323.
Palamau district, capital Daltonganj d - q
Panchkura . - 1 e e 32? 97’ Be", a7’ 133.
Faniathit R, . v e ٠ . al aS ge 91? 20' | 115, 120.
Parbatipur . - . UT x 25° y 88° 58’ | 76.
Patna . e e e Š e 25° 37 85° 13 | 41, 328.
Piploda A AR pie ER ME E 75507 Wf. "Lan So.
Pirgunj : e : e 2 25° 25/ So? 22 1310;
Pundengru . . . ۰ ° è ۷و‎ eat on" Hy (117
Purbasthali . ? e : ہو‎ 234 88? 24' | 324.
Purniah e r . : : 25° 46’ 87? 31 109, 308, 326.
ata Ee e. ek ae So go” 57’ | 330.
Rajagriha (Rajgir) ۰ . . Lies n4 85° 28’ | 338.
Rajshahi district, capital Rampur Beauleah dë SE 109, 319.
Rambrai ° e . e . . 25” -30l 91? 23’ | 81,121, 131, 946.
Rampur Beanleah KN, «41 7242 24 88* 39
Banpopunc iN CSC," 3 as 92° 47’ | 98, 338.

EE

x INDEX,
Lat. N Long. E. Page.

Rangia . e e . e e wi O Heemer 1.524,

Rangoon . . e . e 16° 47’ 96” 13 | 63.

Rangpur . e . e . e|. AS A5! 80” 18. | 43, TO2, 100, £37: T72;
284, 318, 354.

Raniganj œ e e . = ele SE I LO! 37% 81 Mok, 324.

Raninagar . - e - e A en Li So. 2 "1-989,

EE" O ےک‎ A A a ee 75 7 50:

Roman ہے‎ es el os! V DEP A5 OS ID SO SO:

Rü ood us uo T ers DEE 93° 52 | 67.

Rupganj e e E e - - 239 48 gei 34 | 820:

Rupsi + . e " a . E C7 siet sd Sei so” | 19, 127.

Sabhar . e e . . . wot B3 ST. go” 18 | 329.

Sahibgunj . ` . . e D VE 87° 40’ | 309.

Saidpur e E : . e ` "luo rd 88° ei | 280, 319.

Sailkupa . e . e e . 23 41 892 48 (3282

Sakalipur . e . - . o9 کو‎ 87° 56 | 325.

Samin + ^ e e . - : 25 41’ 90? 43 | 147, 156, 197.

Sandai . e - . s e d ے2‎ o d 0274 "25 TT,

Santal Parganas district, capital Dumka ide ۶2 IIO, 193.

Saraghat œ - ۰ e e si^ aae. ubt 89° e | 64, 71.

Sariakandi . e . S - S 24° 53’ 89^ 38’ | 322.

Saugor Island . - ; z 172179217 30. 88" 14” (133.

Sehore oe A رع ا‎ Ee Rea oda

Shaistacang 4x — نے‎ ks S 2 LORI ZA e 91? 30’ | 297.

Shampur . e e . e ^ DEP A2 89? 13’ | 284.

Shella . = ` : 3 : 20ے‎ C 91? 42' | 120,

Shiradize an 4" à X QUE LE 88° 45’ | 39.

Shillong "aue EE ies he CSE 91° 56’ | 4,43, 75, 78, 113, 5124,
126, 136, 169, 171, 178,
190, 192, 209, 267, 313,
345» 356, 359.

Sibsagar : 5 : e E |) Toy 94° 38 | 110, 340.

Sidugiri e e 5 e e 41: 125^. 922" go” so” |156.

WEEN

Page.

29, 78, 81, 204, 318, 343,
346, 355.

63, 71) 76, 282.
1215152;
301, 323.

163.

635 71.

128, 276, 344.
325.

329.

261,

326.

280.

320.

109, 326.

335.

125.

78,109, Ilo, 115, 110,
128, 134, 161, 170, و171‎
178, 191, 274, 343, 340,
357-

39, 194.

334°

331,

9%, 10$, 119, 211, 267,
318, 8۰

39, 67.
194.
282.
310.

110, 332.

287.

—————————M——— —M———— M——————— n nee]

Tipperah district, capital Comillah

Sikhim district, capital Darjiling

Sundye, see Sandai

a

xii INDEX,

Lat. N. Long. E. Page.

Toungoo (Taung ngu) . ۰ ۰ . 18:53 96^ 51^ | 5L

Tara ^. » S ۰ e ` : 255.20 go” 16’ | 9, 84, 91, 118, 124, 126,
127, 136, 159.

Ukiam . ` S ~ e " |. 959. si’ 91? 25’ | 122, 151.
Ulipur . : ` - e . 3 25° 39 89 41 | 320.

Utripura e E e e e e 26° 45’ 80° 10’ 65, 71.

Vishnupur . " e e . D WE EE 87^. 23 | 325

Vizagapatam ۰ E e . > 177,42 83° 20 | 56.
Wokha - - . > : 41.526) - ^6 ہو‎ 19 | 194

. S : 17° y 95 41 | 67.
Yatung > MMC e * : رد‎ 35 88° 56' | 38, 193.

Yandoon (Nyaungdun)

INDEX.

INDEX.

PAGE. PAGE.
Acceleration, of wave particle ۔‎ 78, 344 Bent rails . 97, 280, 286, 292, 297, 338
Aftershocks ; S . a B, 124 Bhorka œ. : e e = S 53
— -—, changes of level due to . 159 Bibliography + vu e کو‎ S 377
— ———; distribution . . . 128 Bickers, A. L. , e . : . 66
— — , in epicentral tract . 120 Bicyclists, affected . . e ° 33
=, landslips caused by 114, 115 Bolton, C. W. . . > ` - 378
, local character 4h oce 26 Band Was wie’. @ up 302 363
Agamennone, G. 186, 227, 251, 387, 379 Bonham Carter, N. . e ; . 320
Agram earthquake, 1880 . . « 57 Bordwar fracture . . . .148
Bion B.E. ٹوو وو ہرو‎ ors eu Borgen, C. S f : MXN TIC
Alluvium, displacement of . - 94, 297 Bose, P. N. e - e کے‎ 2I AI
Amplitude, of wave . . . 82,117 Bourdon, "E Be as). BC ou. 11226227
"Anderson, J.A. . . : í . 98 Brahmaputra, changes of level op . 163
Arbuthnot, J. C. e : >D 156 Bridges, distorted 95, 104, 274, 286, 295,
Aslett, Mr. ‘ à á ¿ . 37 Brownlow, H. H. . em e . 171
Assam range Burrard, S. G. . e . . 58, 361
, fault scarps in + 135, 167, 368 Bysack, Gour Das . P . 200

» geological history . i 4432

Cachar earthquake, 1869

e $ . 85

physical geograph . et
d dies ^ Calabrian earthquake, 1783

. e 212
structure e ۰ ء‎ 165, 36
s i Cancani, A. . + 204, 233, 255, 378
Baratta, M. 7 e " = 234, 377 Cargill, 1 D. L] e . . e. 316
Bardalai, Madhub Chundra . 102, 335 Carstairs, R.

D D e a . 327

Barisal guns . . 200,263, 208, 318 Catania . e e . 235 240, 375

eo seismic origin > E . 205 Cause of eart quake . 176, 179, 367
D

Barographs x / . 58,61, 180 Charleston earthquake, 1886 42, 48, 57,

97, 101
Batavyal, Umesh Chandra وو ا‎ Chedrang fault 80, 138, 145, 197, 369

xiv INDEX.

PAGE.

Earthquakes, Charleston, 1886, 428, 48,
57, 97, 101

=, Japan, 1891 + 3 . 97
~~, Laibach, 1895 e e 57
و-ے سے‎ Lisbon, 1755 e e 371

ES A Neapolitan, 1857 e . 357
o. Riobamba, 1797 ie Ca

— =, Sumatra, 1892

, Vostizzia, 1861 . 86, 100

Edinburgh - e : ë . 238
Elevation, changes of ; see Levels -o

-, of river beds

. 13, 104, 31977
Elison, J- e H e DH e e 102
e. e o e e 323

English, A. E. D e. . . 39» 194

Emerson, T.

Epicentral tract e e . e 129

Epicentre, extent of « e . 168, 172

Eschenhagen E e ^ Uis 242
Evans, J. D e SER. . e 37
Evans, R. | و‎ PIC e 2 I57 271

Faille du midi .
Faults, Assam range + e ۔‎ 167, 369

- earthquake . 5 . 380, 138
Figce, S. + s nis e . e 241
Fisher, O. + : : . e . 177
Fishes, killed by earthquake . e 80
Fissures, at foot of hills e . : 7025
— ——, distinguished from fractures + 6
==—, double - . . . . 203
—, formation of + e e . 89
————, in alluvium 21, 22; 25, 85, 109, 258,

280, 292, 319
———, on hill sides + e 6, 10, 11

Floods e e . ot e e 161

e 85, 296

PAGE.
Cole, H. W.G. .
Collier, F. R. S.
Colquhoun, H. A. C.

e e 28, 342

e ٠ e e 332
. ۰ e 360

Compression, of alluvium + e 107

nn in Assam range e . 366

Condensational waves 5 20239. 2S1

Construction, influence oL 58. : b 44» 304»

I

Craters, see Sand-vents Cryptoseis-
mic e . . . e . 2
Curtis, Eco e e ` e o 65

Cylinder seismometer 294, 345; 357; 358

Darwin, C. e e 27212, 217
Davis, W. M. +
Davison, C.
DeBure, P.
DeSouza, F. S.
DeTivoli, J. W.

Displacements, of Hills

. e . 7ءء‎
‘ . e 124, 177, 223
- . e : SEKR
. a 04
e . . . 98
^ e 165, 363
. 82, 297

e e e 232; 252

— —— OË Alluvium
Distortional waves
Dobson, G. T. H.
Duggan, C. R. +

e : e 15, 107; 127
AS e gn OF

` 5, 6, 9, 16, 21; 26, 205
9 31, 33, 34> 38

` $8 ھا‎ 42, 48, 57
7 e e e e 36

Duration of shock

Dutton, E.
Dyer, J. A.

e . 163, 369

Earthquake, cause of
سم‎ Garo explanation کے‎
= Hindu prediction «e 24

7 Hindu view of cause e 24
Earthquakes, Agram, 1880 . 2 57
E EE 1

— — y Calabria, 1783 2 2ء‎

INDEX. xv

PAGE, PAGE
Floods, caused by rise of river beds . 106 Harris, T. . . . . 0ء‎
——, —- landslips. . . 121 Haughton, S. + e . . . 346
Focus, depth of . e :۲375ء‎ 788 Hayden, H. H. . e 2, 325 00 1172277
=, observations to determine Heath, T. e e e 238, 377, 378
ہے‎ E Se را‎ ee متا‎ ere PNE MEE
——, SIZE. ۰ 165, 168, 172, 179 —, rent . . ۰ ۰ . ° 149
=, thrust plane . e e 166, 369 Hobson, H. G. . . e 6 e 38
Folk lore . . Ap e 14, 24 Hodograph ۰ e e ۰ .-73
Formula, amplitude . e . ° 349 Hot springs e gum, e 41, 328
o, acceleration . ۰ ° 348, 350 Howell, A. A, . . + 9, 130, 156
——, velocity . e . e 347, 351 Humboldt, von . e ew 8

Fracture of stone posts 172, 271, 317, 318

Inglis monument . . ` . 208
=, of trees . e 81,123, 138, 150
Irwin, J. R. " ۰ ۰ . e 66
Fractures e ۰ . . a D 138

Ischin ۰ . . . D 2:228

, distinguished from fissures +. 86

Isoseists . . e e (10424: 49
Früh, Dr. . ٦ H . B B 237 1

Italy, shock felt in . ۰ e e 375

Gabbett, E. : e » = «4.30 Japan earthquake, 1891 . e e 97
Gait, E. A. . . a e ° 378 Kemlo, G. e e e e e 97
Geysers’, see Sand vents Kennedy, W. M. e e e , 341
Ghose, Nishi Kumar . . e a) 75 Kershaw, L. J.. ۰ ۳ . e 343

Gilbert, C. K. e . . H 40

Lac colite . . . e . e 168
Gilbert, R. H > > . . 33

Lafont, Father + e e e . 30
Grablovitz, G. + . e 228, 250, 251
Laibach earthquake, 1895 . e e 57
Gray’s theory of rotation 215, 218, 224, 6
Lakes, formed by faults . 80, 139, 148
Grenoble , e e . . e 237

y lan dslips e » 121

Grey, M. A. > " . E . 338

warping e e 152

A A 7 :
: Land-slips e 6, 10,44, 111, 119, 274,
Grimes, G. E. . . * 2; 91, 04, 288 331, 334, 335

——, caused by after shocks 114, 115
Guha, Abhya Sanker . . . + 336

————, due to height of hills . e 113
Gurdon, P. R, T. » 27, 80, 99, 100, 102,

334 =, effect on river channels . 120
Hallifax, A. G. 4 è É e 333 ———— , floods caused by e e 121

Harris, E, B. - . e e e a 221 — —— a lakes formed by e ° 121

svi ` INDEX.

Paez.
Moyle, G. 1 oa TR . 98
Mozumdar, Kedarnath . e . 76
Mud volcanoes A : Lo wd
Mukerji, Hiranmoy
Mukhopadhya ya, Pyari Lal Ug gree

Napier, e E E e A 35
Newbould, B. B. e بی‎ +. 343.
Negative rotation . . d . 208
Newport . e . e » 238

Nicholl, B. V...

Oddone, Ê. . .
O'Donnell, H.
Oldham, T. . es ےا‎ enn TAS
Oldham, R. D.

Origin, time of

. . e 317, 377

e e . . 75
Overthrow, direction of . - e 1355

Paulsen, A. . è - - e 243
Pavia . P > ^ e 236
Pereira, P. D. C.
Plax e Om : . "SEN مر‎
Porteous, A. .

Positive rotation

Potsdam

. e ° 208
e . . D D 237
Preliminary tremors. . 241, 247, 251

Projected objects . - 262, 264, 290

o. stones ^ Á 11, 78, 130
Lee a limits of size. e 132
Ree و‎ overturned . ى۳۴‎

Projection not due to wave motion 82, 353

Propagation, rateoffeltshock 53, 70, 74,
248

, unfelt shock, ist phase ۱
247, 5

PAGE.
LaTouche, T. D. . 2, 96, 178, 257, 377
Lea, J. H. : - . Tak: Y]
Leghorn : ۰ è ` (^ Sy

Level of nostrain . i e 7
«e 162, 363, 367
14 157, 27!
——, — caused by after-
shocks . e 359
, disturbance of . . . e 05

Lisbon earthquake, 1755

Levels, change of .

e ۰ ۰ ۰ e 59
e 3735 372
Lussana, 8. . : < . 236
SE 3, 378
. . e 25; 91, 99

Luttman Johnson, H.

Lyall, D. R. e

Macintosh, H. J. . . e «05825
Madhupur Jungle . . e 09, 292
Madras time . e - 5 - 56
Magnetographs . e e 80, 241
و‎ cause of disturbance

181, 185, 189, 245
===», value of records e 245

Mallet, R.

81, 85, 89, 212, 218, 224, 226,

l 357
Marina . e . : - 204
Maxwell, H. SE, P.

Meleda,

€ . . 103
D e . . e 203

Men, stunned by shock . : 22

——, drowned . . . 80, 142
Mercer,F. . e 1 e e 20
Milne, J. +» «© 74,133,238, 250, 377
Misipeffers . e . pn Zon

Moos, N. A. F.
Morgan, J. G.
Morgan, R. R.

e - 38, 180, 378
e e b 190, 192
. : 102, 163

Moureaux, T. - - a 2905 377

xvii
PAGE,
Shillong plateau, see Assam range.
Shuttleworth, A. E. : > 26
Siena. ٠ e . . 609ھ‎
Sitakund, hot spring

Smith, E. H,

D . . 41

. . o . 4
Sounds, earthquake ^ 4, 6, 30, 191
— -, heard under ground . ۳۷ء8‎

——-, explosive. e = 28, 194, 208
See Barisal guns,

Spinea di Mestre 9 97 . «1378

Stevenson, S. e - e « 324

Stone pillars fractured

Stow, Mr.

172, 271, 317, 318
D L] . o . 60
Strain, level of No. : $ acid d.

preceding earthquake . 167, 369

Strassburg

. . D . e, 237

Stratton, W, à ` : e 39

Subsidence, between fissures . 5 91

-———-——-,ofriver banks . . 106

Suess, F. E. - e . . 57

Sumatra, earthquake of 1892 et Se

Sur, Devendra Bijoy e e e 102

Surface waves 5, 6, 20, 26, 27, 36, 37, 40;

253, 334, 335

a etm - amount of tilting 34, 38
سے سس‎ height o 7, 227
— — Es length D * 73 37» 227

-, velocity

27, 248, 253
Tanks, filled up . A 18, 104, 3194

Thrust plane . ` - A ll 7366

Thurloe, ]. W. i A ` «30; 64

Tide gauge . = . : «57, 63
Time, Standards in use . . ^ 55

—-— - in Burma

Trees killed by shock . > 141

INDEX.

PAGE,

Propagation, unfelt shock, 2nd phase
248, 251
pe ——-. ard phase 248,
253

Radice, C. A.
Rails, bent

hu A . ء‎ 3
97, 280, 286, 292, 297, 338
Rajagriha, hot spring. 0 41, 328
Rebeur Paschwitz, E. v.

Ricco, A,

. H D 25!

D . H . ۰ 235
Rioba mba earthquake . e st چوس‎
River banks, subsidence of v £06, 161

——- beds, upheaved . 13, 104, 3٢١9

Rivers, rise of. e " 4 107, 162
Rome: . a . . b . 233
Rotation. ; 207, 260, 264, 273, 283

St. Dalmas, H. de . = e e 37

Sand-vents . 13,15, 258, 285, 292, 319

=, depth of . . « 322, 339

——-, distribution of . . . 109

—=, formation ot. . ۰ ۰ 87

——-, formed after the shock . 16, 20, 25,
99, 327, 334

15, 20, 25, 26, IOI,
103, 322, 331, 335, 337
-——-, temperature , e . . 16

——-, force of issue

Samin fault . ^ d e EY

Schmidt, A. . - 73, 173, 176, 250
Schmidt, J. F. G. e ٠ 86, ٥

Schollengebirge . . e 167, 368

Seebach's hyperbola : e 92 173
Seismic area . a e . e 42
—— ——-,extent. o : و‎ 52
——— >, outlying . e 50

, in Italy : e ا‎ fe
Seismometers, cylinder 294, 345, 347, 348
Sekiya, S.

Shide

. ۱ H e $ $ 133

. D . D ۰ 238, 256

xviil INDEX.

PAGE.

Warneford, R. e e . 36, 65

Waterfall formed by fault a 130, 141
Wave, acceleration e |) 98

amplitude e D e 82, 117‏ — سے ۔

ageet? height . . e e ké 227
——-, period . ^ - e 8, 35
—— -, velocity of particle . 81, 343

A سکس‎ propagation 53, 70, 74, 247
E 37, 1334
A I8, 104, 316 fF
CAR NS rn 348

———, visible

Wells, filled up
West, C. D. à
Westmacott, E. V..
Whiten GSG. ai «x e 7. 66
Williams, S. M. J. .
Williams, O. O. . e e e 29

Wylie, H. 4 e . a e. 3 8

`

PAGE.

Trees overthrown. e e > X18

——- broken by shock 81, 123, 138, 150

Trigonometrical Survey . e 138, 361

Turner, de W. © P D H 35

Unfelt eartbquake ` . 417227

Vents, see Sand-vents,

Vertical component. e . 120, 352

Viuntini, G. e e e e € 234
Violence of the shock . á e 78

Verticose shocks . e . e . 240

C. Darwin on. . « 212,217

C. Davison on + . ewes

Vostizzia earthquake, 1861 . 86, 100

Vredenburg. E. . . - e 2, 302

Wahner, F. +

020 670 20 8 LNIYS E

y ddey y‏ 9¿ مرہ:

osang- Sugod

"opt Tudy eynoreg'seoyzo epu] jo 4
J ۱ 3

(II
M VIE ۰۸ء"‎

mg 7‏ ہیں
A A sp : kc :‏

T'Id YIXXTOA 7 3 "UrEWPIO (T W

"VIGNI JO AdAuns 'IVOIOOTOHO

HOAN وو‎

"8681 Ley e

0

00/29007 erpug jo ÁeAung Suan- g

A E?

04d ¿9 3 ddey y

k : ^ E Sy Se
io 5.
y EU. À
ES Kë * , aN : -
و ںا‎ Y مر لے‎
H Ew - i ki 7 " ’
y 1 ۱ à; EX
"IL Td XIXX PA sro ure AY 4-167٤

i‏ ر0۴3 0ا19 ۸34169 9000 ۱0,99 :9 ۱۵۱ج

ANOTHAS SIVE 100060-00 07600777

? ojoud 900090, ERST 1

2681 م‌ممجدہ' رھ‎ ‘SFO erpug jo Lonne - Suryoye-oj0yg

“TL 1d XIXX PA 800006۳ "UreupiO (TH.

"NIGNI AO ۹۹ 808

۰۰۱۱۷۰۰۰۷۰۱۷ wA Hao uS 08

‘oqoyd næse Jeu) yr S

d *

1 key’ EVMI 5'seoujo erpu] jo Aeaang -Burqo39-o30qq

“A Td XIX TOA Sato 00۲ “91010:07

"NIGNI dO AHAHAS TVOLOOTONO

IPNDndVHHHHO ۸0107100100 NI SENOL

“S681 UNC LENNIE D‘ S200 erpuy jo Laang - Suryaz-oyo q EE

7

/

-ureqpto YH

"AC G UE O STOUT

"NIGNI JO ۱ 0:031 TO HO :

۸۷۸۷۱۷ ٦ VHVdTIVH -UNAZAL VEVIVINVA LV STIVH 68

DEI مچوددہ' جح مممو>وصہہ‎ erpu] jo AaAang Sms oo "ootd ema
ER vie, 7 KC Cup |

IA “Td XIXX OA Slowey

"Ure qpro' TH

"NICNI 4O AJHAYHAS 'IVOI2OTOJO

Memoirs Vol. XXIX Pl. VIL.

E

June 1898.

Photo-etching-Survey of India Offices, Calcutta,

EE e SURVEY OF INDIA.

a‏ > مخ

MAN SHAT BRIDGE.
EASTERN BENGAL STATE RAILWAY.

R.D.Oldham.

PECH

AVM'TIVE A4LVIS TIVONHS 08781
EEO Vita ENEE NO DY LS

LU

'gegr eunp^ennoqpej'soogjQ مھ‎ jo Zaang Stong gou 'o301d. ¿9 3 ddey 3

5

je Nt G‏ سم
r mI a d ng‏ ہبہ نجج جت -

“TIA "Id XIXX TOA 860۷۲ 6

ء۰۴ ۰۱7700110000

1 AVM'IIVH TYONHE -NVESV
NOILVLS CNVOVLSIVHS EVAN SLNEA ANYS

2% ddey 3

: 8681 8008230008 028000 erpuy jo Keamng Suryaye-oj0yg
z ا گی کے یں نے‎ ETA S TA d SCH کا‎ SE

€

'o30ud y

"XI ld XK PA steet -Wep ۳

VIGNI 40 ۸70۸09 1۳۲70071009

GEOLOGICAL SURVEY OF INDIA.

R.D.Oldham. Memoirs Vol. XXIX Pl. X.

T.D.La Touche photo. Photo-etching Survey of India Offices,Calcutta, July 1898.

"07۹ MART.

ZA
Ce

Kë:

t

0 A ٰٰٹ‎ َ 70787۵8

gear Af emnate مأ رود ٭'‎ erpuy jo fang -FurU218-010U f :o30yd' ayonoy, BFT J

AMEN کر‎

ES

TX 14 KIRK eh 8-071 | 7

“VIGNI dO 01ص۸۳1۸‎ ۳ 0۲80۳۲0 ۳ :

OM. ONVTIVT LV SINENANNON VISVHY
OM ONVTIV'T LV SLNSWNNOW VISVHY

V HZ
f -2yonbuyyno mq) fiq- peddwus sypouou fo seonfms uryosg جح‎ o
7و وو'‎ “BROTE ‘S9380 pu] jo Sang ۱ | m

E
. MAGNI dO AMAS 1491901043

fe pe SS esM اف نے‎ "E کی سارہ‎ Mu AE تس سے مہف ےر رہ تی‎ tu A ور ہو‎ 1A eS e سح بے 2 ن‎ > E" p " * "` ei aiino d
= n a” T ` >i - 3 d

AIGUL TO YEVAUVZ JADIOO JOAO

ve, A ; E ADCIS XDOCIGV. aaiomoM- y ^i . ae x CR H w e te ۱ 7 : | ay > | 0 - | | | A i ; | a سس‎ FR 7 ۳ s E : .medblO.Q.51 E
$ > 1

. DAL oN yá dos apar, BAYS onam ”ما‎ zoo wi» این"‎ "a o»

TON ÐMAJIAJ TA @GTUAMUWVOM AIZ AHA

4LOM + 7 7 1

$ 600 eunt “833101-080 ۶86800 Epu] jo ۸8 i guya =010Y Y
ag 5 NRT. RAE " Toe ہہ ڪڪ‎ ^ ` " ^ " 2 = - = E

ER x *‏ کس
A e T. Em‏

ر
.^
i=‏

2 p.t
1 7 Ee ` Ea
T SC? "3
رڈ‎ d
M D
5 j تک‎
IX Td XIXX TOA ٢٤000 UrewRIO Cre

“VIANI "۰080-۸ 4٤۴ z

As. oe تر‎ >

pu
Py ^

YRS

"t

۹
۹

emie: ADA doro Ai. 1242] fijtouriof acom 2/204 Jo sdoj 2) "asnow oron po fe SPOT RR. jou

001وت“
انت

y گرم اتا‎ a

D

لاق

H
2

e
Qu
"KS K

d

1

ON

d

AN doo

d

۸3908.88

٦
>

a

٦

N 48

NTAV

SHIHE ouo

vid

Vi

t ,
MKIA AKT 8

d

cR

MIX XI LOV ierch ` ` | amadb[Q A3:

we ens vw Sova) ۹۵ہ لپ‎ anaw ossa Yo ego S sawol ov) blo "o 226% o.w

ا MIMAS TA CAOA AHT ٥0188070‏ 07۸0 11ل

STIIH OYV9 NIAVS LY AYON HHL 9NISSONO LINYA

‘9691 eun, “800060 5070 VPU jo ۸)6

SUYOS - o10 J

THX Td XIXX 19A 00017٢ wepo au

VINT dO AHAYNAS TVOINO IOUS

7 SA
SCH

15977 ONVHGaHO FHE AF 8800187 0

waste og fig ybno.ng qna rf mm mo, _

یت

WHEN 9اد‎ FANOS 1۷00۵010

jo 69

1
d

empu

hoy

5

870

210

ار

'gegr'eune 7

Xo

/

ANGEL

۹

MIX. IT XIX X .]۷ ٠110114

p ex

d 2 Ze “a

i ' ~
AN 5 < ,

Aw F. ‘ 1

` 13 ` Ca

TIUAY DVAAGHHO HHT YA ))7 ۷۷

1٦1۷78 ONVUCHHO AHL Ag CENO TIVIHWLVA

'gagr'eunp “77077 'seogg() epu Jo Asaanc 'Surqoje -oyoq d

Mur 5 ^ PS >” a

AK TE XIX OA 7 7

"NIGNI HO -AJAMAS 'IVOIOO' TORS

Rutt

3

TEAL ONVaACd ONG 8 اناو و و‎

LAT QHHO HHL 40 qud AH. + LO y VAJAH dE
‘SAATE ONV + quHo JHL dO Gad ¿HL ONILLOO LINVA ONVYC: HO JHL

ATW
IN Y
LN

۰9601 eun "exinor 7 9000 SIDOT jo ۸۸.8 9Urqoje-o10q d

6 ۸0ج
يہ -د7ء٭م

do

DTE

"VEO 5907000027 TAI PONTOD

2

a Lee oo لکوت‎ qe

zi —

.AIQMI YO YAVAUZ JADIDOJORO -

Wy, d
TON.

hores te
:فدعط‎

AQVIS OMAAIAHO JHT- YO CAA THT OMITTUO TIUAT DUAACAHO FHT

SAN ONY ACHRO HAAL 7000070 0 ONV 701

'gggr'eunp'exmope) ‘S2030 TPU Jo عسمھ‎ ‘Suyana opoy Y

ا

‘AX IA XXX TOA Sowa ات‎

SLOPE FORMED BY THE CHEDRANG FAULT NEAR DILMA.

Survey of India Offices, Calcutta, June, 1898.

Photo-etching.

Memoirs, Vol. XXIX, P1. XVI.

R.D. Oldham.

GEOLOGICAL SURVEY OF INDIA.

GEOLOGICAL SURVEY OF INDIA.

R. D, Oldham,

Memoirs, Vol. XXIX, Pl. XVII.

FIG, 1. ALL SAINTS CHURCH, SHILLONG,

Survey of India Otices, Calcutta, June 1899.

2. OFFICE OF THE SUPERINTENDENT OF TELEGRAPHS, SHILLONG,

da i7 ^

[^

s af Tyrii nien d SE

arem و‎ Se بت‎ S il I ام‎ E

——— —

-——— ip ©

m

D

4x de
P م٦‎
H 7
Lë
D
ا‎ a
^
ru > 2
d n
- v d a ۸ z D
KA y E 4
نپ یو سم سم نے‎ Antic ^ oh یں نے‎ TR SEH
LE
see
k D
y `
Ste
ری‎
ws
Us
SES

GHOMLOGICAL SURVEY OF INDIA.

R. D. Oldham. Memoirs Vol. OI 2

WM جک میس‎ TA
: 3 E

FIG. 1. GOVERNMENT HOUSE, SHILLO?

(Before the Earthquake.

Photo-Block. Survey of India Offices, Calcutta, 1898.

FIG. 2. RUINS OF GOVERNMENT HOUSE, SHILLONG.

GEOLOGICAL SURVEY OF INDIA.

R. D. Oldham, Memoirs Vol, XXIX, Pl, XIX,

ES > s ¡ae
AA ES

" Wi
Le -

: Ai >
e نمی‎

FIG. 1. QUINTON MEMORIAL, SHILLONG.

Phcto-Block, Survey of India Offices, Calcutta, June 1898

FIG, 2, THE SAME, AFTER THE EARTH QUAKE,

-
] -— کدف‎ e EC rli
^d Pd. Soa y PLC. sat 4
pe. P
á + peu Vm ہ ےہ ےد‎ apte
— e
,
a ۴
=
e i
à
p H
, D
H d ~
^ t
E. d
> 1
d i
B A
y
E E
7 i
H
۴
| 7 E]
" i
2 D
P
gy =
H
E
[4
7 >
a -
3
A
y
1 d
E E H
i
* =
i
;
3 -
A
H
1
$
E D
p:
i
H
se J
»4 Due [D 7
a 4 1
: el Fd
- + i +
3 : a4
رات‎ 2 s ,
AE
bou NE
ies
LS ve
E Dos
e x
* :
E
> H d
i
H ^

MER O جج کہ ےر‎ Ai

Ne ER
می ہہ‎ IE YS
DU و می‎ ES

GHOLOGICAL SURVEY OF INDIA.

R. D. Oldham. Memoirs Vol. XXIX, Pl. XX,

FIG, 1. DEPUTY COMMISSIONER’S CUTCHERRY, GAUHATI.

Photo.Block. Survey of India Offices, Calcutta, June 1898.

í n

ey‏ مسر کے کی اک ایب تو
e sth‏ یہ ےم

>. ا کا ایا‎ ET ا‎ :
A ETT ARM Lu :

DENTATIS O A. aoa pr IN AMO ۶+

"S

Û ae Aet reg" n

geng. deg

۷۴۰

+ بس میں 1 ھھ‎ ae
Léi

—

ILLND NU a ME I AT.

J D
TE ARA اسب رب‎ Eos diris em ی‎ ae wm وم‎ ma تع‎ > mo «mma نے‎ et ےس ام‎

SURVEY OF INDIA.‏ یئ

R. D. Oldham. Memoirs, Vol, XXIX, Pl. XXI.

No. 3 NO. 2 No. 1

FIG. 2. TOMBS IN OLD CEMETERY, GAUHATI.
(See also Plate XXXVI.)

Photo-Block. Survey of India Offices, Calcutta, June 1899

AE FOE‏ یں ا

61۳701) SURVEY OF INDIA:

Memoirs, Vol, XXIX, Plate XXIV.

FIG. 1. GOALPARA BAZAR FROM THE EAST,

Survey of India Offices, Calcutta, June 1899.

FIG, 2, FISSURE AT HAST END OF GOALPARA BAZAR,

R. D. Oldham,

Photo.Block.

GEOLOGICAL SURVEY OF INDIA.

R. D. Oldham. Memoirs Vol. XXIX, Pl. XXIII.

FIG, 1, 9, HARRINGTON STREET, CALCUTTA,

EE EE e o

E y

Photo- Block. Survey of India Offices, Calcutta, ۰

FIG, 2, 6, ELYSIUM ROW, CALCUTTA,

EE‏ 0ھ
ade FAL KÉ ACUTE‏

LÀ -‏ کہ ہے
DIT 7‏

MÀ Áo‏ سے

= 1 z

x i

۰۳2

Ts SA

=

H

H *

t

sono MUR rta orm mma s men
Us
۴
H -

dip am t
A
D

D

«merme un

BZ
Ge 7
eS ele ہے‎ E

EE SC?‏ سم ces oy See‏ ہے

1

rl + * >
Ka? SC KI
HI Kee e "-
Ch ds

Ki
t دی‎ d
y A
7 3
Z

iE

ij tas Ps
5

D

n +

1 e
i27

E

:7

1

p?

e H
23 :
p

D

Sch

pe

i

te

it

AS

Zä 5

1

01

3 Vi
7 مت‎ 3
4 : x:
ist us
jt M

Es

D

ER
"A
e,
M
2
D
7

[os
|

D am 3

Sec با ہے‎ E

p: ہیں‎ YE 3 e
n
pu
Des
e 8
H
5

D

GEOLOGICAL SURVEY OF INDIA.

R. D. Oldham, Memoirs Vol. XXIX, PL.XXIV.

FIG. |. No. 114, CIRCULAR ROAD, CALCUTTA.

Photo-Block Survey of India Offices, Calcutta, 1898.

FIG. 2. HOUSE IN BENTINCK STREET, CALCUTTA.

D
7

(P وہ سی‎
H

——— ap

H

پ ہی درک ہہ
j‏

aere m Daer

y GN t a s,‏ دہ

"ud - y Je M
` H wë, dech RR E
Ww ہے‎ A Ware MH eee nn mp mi me AA e ت سے‎ enn me

CHOLOGICAT, SURVEY OF INDIA.

Memoirs, Vol. XXIX, Pl. XXV.

R. D. Oldham.

FIG. 1. FISSURE IN COMPOUND OF JUDGE'S HOUSE, RANGPUR.

Phcto-Block. Survey o1 India Offices, Calcutta, 1898

FIG. 2. COLLECTOR'S HOUSE, RANGPUR.

GEOLOGICAL SURVEY OF INDIA.

R. D. Oldham. Memoirs, Vol. XXIX, Pl. XXVI.

FIG. 1. BRIDGE OVER STREAM AT RANGPUR.

Photo-Block. Survey of India Offices, Calcutta, 1898.

\

FIG. 2, BRIDGE OVER CANAL AT RANGPUR.

GHOLOGICAL SURVEY OF INDIA.

R. D. Oldham, 6001002007

FIG, 1, FISSURES IN SANDSTONE AT EDGE OF THE BALPAKRAM
PLATEAU, GARO HILLS.

Photo-Block Survey of India Offices, Calcutta, June 1899.

FIG, 2, LANDSLIPS IN EASTERN GARO HILLS, SEEN FROM BALPAKRAM.

GEOLOGICAL SURVEY OF INDIA.

R. D. Oldham. Memoirs Vol. XXIX, Pl. XXVIII

FIG, 1l, RUINS AT MUKTAGACHA,

Photo-Block Survey of India Ofices, Calcutta, 1898,

FIG, 2, HOUSE AT MUKTAGACHA.

GEOLOGICAL SURVEY OF INDIA.
R. D. Oldham. : Memoirs Vol. XXIX, Pl. XXIX.

FIG. 1 GODOWNS OF JUTE MILL, SERAJGUNJ.

FIG. 2. MANAGERS BUNGALOW JUTE MILL, SERAJGUNJ.

GARO HILLS.

ALLEY,

r
P:

d
O
2
«
E
g
P

d

RRAPUN

HE

d

C

AR

NE

IPS

DSL

LAN

1

IG:

F

x r

RIO

۸5

XXI

irs, Vol.

Memo

Oldham.

R: D.

INDIA

۸0 OE

N

AL SUR

EOLOGIC

G

۰٠٥90010 '(qUH

IHONVM UVAN SUACTNOG 1

HO AHAYHAOS "IVOIDO'IOHO

t

IS VHM

SE

GEOLOGICAL SURVEY OF INDIA.

R. D. Oldham: Memoirs Vol. XXIX, Plate XXXII,

L e ——— — ee

E i
Le
rct unii I

—À—— A e

Pa

FIG. 1l, TOMB IN DARJILING CEMETERY.

(Top part twisted).

CI SERE. i
M ارك ا‎
d af CT

KIM
AL

|

ا
mei‏

ات سح |
Pads‏

ا
qa‏
i il 0‏

R
al
|

| i uidi d
1 i AENA ا‎
| LI 1 | ii : HE at 7
ا‎ A H ¡A
Beg! Mg 5 وا‎
7 Li
tilii. : +

000

3

Photo-Block. Survey of India Offices, Calcutta, 1898

FIG. 2, CRACKS IN WALL OF JAIL, MAIMANSINGH.

EA
a

CH
1
Mi
H
»

eer. Séile 27
Tw m Ok

GEOLOGICAL SURVEY OF INDIA.
R. D. Oldham. Memoirs Vol. XXIX, P1. XXXIII.

H

AS
EY
i

200

Plan showing position of 223% 4% and 5% blocks
from top of obelisk.

WILLANS MEMORIAL, SHILLONG.

TE ای کیو و‎ FIRES "
. شر یں‎ IA ٹوو‎ P
: ید‎ no
j ۱ A SCH

d

GEOLOGICAL SURVEY OF INDIA.
R. D. Oldham. Memoirs Vol. XXIX, Pl. XXXIV.

ol
" f

AN

KIK d q 0
j

Ht {pes Gel ea has DN os Ae

S. pilar fallen due N.

AAA ج‎

E

I
E pillar (wusted |

Hastern par

TWISTED GATE PILLARS, SHILLONG.

*

09

*

GEOLOGICAL SURVEY OF INDIA.
R. D. Oldham. | Memoirs Vol. XXIX, Pl. XXXV.

Monument to George Inglis at Chatak.
(See Frontispiece )

Fractwre

Plan
Twisted, B

Portion detoched

-

P a |
1.5“ Aqueduct Pillar

Pilar on wall

Twisted Pillars at Cherrapunji
Inglis Bangalow.

TWISTED PILLARS AT CHATAK. AND CHERRAPUNJI.

GK: e: Weg, Se
AA pM A کے‎
Je pr IR

yo

eg `

P رہاب‎

Se

x

7
لے جج

و

"m.

PU

m.

۶
کو
7

EPs

- : GEOLOGICAL SURVEY OF INDIA.
D? planan: Memoirs Vol. XXIX, Pl. XXXVI.

Memoirs Vol. XXIX, Pl. XXXVIL

West Pillar.

Quarter’s

W. Gate; W. Pillar.

East Gate.

| Signaller’s

t

|

)
GEOL()GICAL | SURVEY `

R. D. Oldham.

East Pillar. — West Piller.

Pl. XXI Fig ا‎
Hust Gate,

Scale I=.

Telegraph Signaller's Quarter’s

f
H
`
$

m '
.
H
D
d o ET
tad
ps * à E m ZUR E
B e نے‎ M a 5 A 4 Aer A KEN AX. T BA"
E 9 á E e é 5 5 La " d $ A EN > » » z
E ‘ » t = 1 A e apa -ET Ed. NA D
be: à 0 H m
E .
g و‎ y E : .
y e z w 27 ھپ تا او‎ gl b 4 a OR یمم‎ y B
^ e 3 " b M Ad E ابو‎ - h Oh KI 3 ER $ BEE H ALT -—
> E T Ki > 3 "

Elim e —
e RA AR a a سس‎

Memoirs Vol. XXIX, Pl. XXXVIII.

R. D. Oldham.

ا
i‏
:

LI

لاج ہے سے کے کے ج ہہ جج ے ے

ooo.‏ وٹ ٹب

wate ene مہم مہ‎ S

7 lé >

Se so کہ کیہ‎

Pl, XXXVII.

y

Memoirs Vol. ٣

SURVEY

GHOLOGICAL

350

40
SS
30

| > رم 1د
| | | | | |

Rr ا‎ A pa aoe
| | “orp (Y | | xm Cra E |

Guzp kno qi SE E کر ہج سس سر ہے‎ --

Memoirs Vol, XXIX, Pl. XXXIX.
100

| CR
EE EE BR
| AGERE DEA پچ جج‎ A TES PME سو‎

1100

R. D. Oldham.

|
| |
|
4

/200

—fioquog

te

او برق دی
کیٹ opungpoon V — SOSNA‏
l |‏
e‏
im‏ جو سس نے شی
|
D‏

LJ
0 1

CIC Pr

BIETER Ln Ke ^ à

Memoirs Vol. XXIX, Pl. XXXIX.

Hodograph.

AD. Oldham.

NDIA.

I

SURVEYS OF

GEOLOGICAL

Libet an Ra

EM ۔‎ Wgl

GEOLOGICAL SURVEY OF INDIA.

"

R.D. Oldham. Memoirs, Vol. XXIX. Pl. XL.

Survey of India Offices, Calcutta, May, 1899.

AT BOMBAY.

Eig l. Declination.

Fig. 3. Vertical force.

RECORDS OF THE MAGNETOGRAPHS

Photo- etching,

=

Memos, Vol. X XIX, PL XLI.

^ KM یو .. ا‎ E ae چا‎ ES لہ ور نو کٹ یں‎
OGIOAL SURVEY OF . INDIN

Oldham.

Memoirs, Vol. XXIX, Pl. XLI.

J2 45 58

i
D

کٹ

ITALIAN SE SMOG RA RALES:

b
ga

GEOLOGICAL SURVEY OF INDIA.
R. D. Oldham. Memoirs Vol, XXIX, Pl. XLII.

ROUGH PLANS OF THE SAMIN AND CHEDRANG FAULTS.

(Figures represent throw w feet)

EE
DIU

GEOLOGICAL SURVEY OF INDIA.

R. D. Oldham. Memoirs, Vol. XXIX, Pl. XLIII.

TRIANGULATION CHART

THE‏ مہ

KHASI HILLS ed
AUMTER H.S.

d H
^

Scale finch = 8 nues. `

Season 1897-98

AOPHANU H.S.

Sriang
h.s.

M OINGHEI H.S.
SF ONOI H.S.
LL ON
MAUTHERICHAN H.S. 2 ON (Laiébti Ae DAR R Sornullor R.s.
` y 3
DN-— A -
٭‎ Nongstoin À y SAilloreg h.s.
A i
| KLAIDERA H.S. Wd
/ | 2 Se a
[| 4
Marskuin 9 نے‎ ongepag
Laimarkham A.S.
- t?
Nongsingri AS, o E Bue u.s

MOSINGI H.S. A

e Mosingram
q THANJINATH H.S.

la Cherrajpoonjt

ا

۲ RANGSANOBO H.S.
MOPEN H.S. 4

MP: engar Baxar

ps
کت‎
1
Ly
-à
EE
bd
سس‎
N
;
Kä
zb
X
ç
à
A

w

1
AK HANDIGAON T.S.

TARAMUN TILA H.S. CON M

™ Chattak

Reg. No. 681, Geol. Sur.--Mar. 99.—1,000. Photo., S. I. O., Caloutta.

T

Arahi 10 پ10‎

pincas, Val, XX1X, I. SEU,

lI —
|
|

| SURVEY OF INDIA.

Memoirs, Vol. XXIX, Map No. 1,

e

Litho., S. 1. O., Calcutta’

I 8 LANDS

of the second and sixth ¡soseists.

The boundaries of the pale and medium tint represent the actual course

are diagrammatic and represent the probable course of the isoseists on the
supposition that the surface was a plain of rock of uniform composition,

NOTE, —For a definition of the isoseists see Chapter Ill.

The lines drawn on the map

Channel

Coco

— — Second
co Wl
—..— Fourth
کے‎ Nit

»

= First isoseist:

Area over which the shock is reported to have been felt.

SS

Essense} Area of extensive damage to masonry buildings.

82 probable limits of epicentre.

Uy

BOMBAY

Moi

HELL,

, & ISOSEISTS.

, EPICENTRE

SEISMIC AREA

of 12th June 1897
SHEWING

EARTHQUAKE

THE GREAT INDIAN

, Map No. 1,

XXIX,

Memoirs, Vol.

GEOLOGICAL SURVEY OF INDIA.

P R. D. Oldham.

—
A
ic

"
s

ou
B
7

- LI ae
Se D e ni ptt
D
٦
D

f >
H $
; e
" iF , es i
B 4 "P ^
IE
PA d
2 a al ty
P. E wen , y
4 i ‘ene D "e
e d E می نے‎ `
LJ e
ay ۱ /
4 7 Ge j
۹ e e a
H e
y ` + ` ۹
E " ۰
ا‎ o: !
t ^
۱ e
| E j "m n
i Se "28

ود

سو
۰

Abee emt age igi a geg e hen Î
NEE ہا وس ہے تھے‎

Memoirs, Vol. XXIX, Map No. 2.

aer:
RON
DE T leo ri KG 7 Singimari ogi a
y Y ایا‎
M "رو‎ pee 776

T TP 3 E — CH‏ و

de Ne 8 Yy y2
lian ae E
Sor WS WU SE y

Ss. Um sanar | Gis SA E St A

0 Zu Zë wf, E ES

> RVA ا‎ Ds

=
SS

1

M da » AG al E e n
LITE SEARS ASAS ES r NES
Dt, ee el ee d Klee

Za ACTES]

: i E ^t
E Y m D £
i j i بے‎ i کت‎
3 و ےم‎ i a ~ e
f n - ` h
€ . M wë
d ` -
5 d VP A -
ul >
y VA D
۹ X x m
y ۲ È E »
i Y " e
5 - ¥ " Se
k
£ E 7 ۴
r B ^
Ce w
7 ye
,
Y D p a
e
۹ D
H
D 4 .
" y

#

GEOLOGICAL SURVEY OF INDIA.
— R. D. Oldham.
S

Memoirs, Vol. XXIX, Map No. 2.

a
ES 0719:1777 T
/ £ aber dotati fe EE al A d
a); 5 / 1 o
Hajabhathichawa Y

Wom; /‏ پیا
Harada ari S x ) X ۱‏ 3
e N‏ ا و gaon e‏
دی re A Gati Khagrfibari E pa Kylaigaonte Kakali‏
P t)‏ : لے ES‏ $ -
G 0 y Sidli Le Ween Kordeguri‏
ا à‏
lc | sp] 2‏
n 7‏

ph $ - Rl V &
^ — B Bara a y
finn o IR 6 |

/ Sy
y SE H SS
` Ja BS 7 7. e
2 > e ما‎ d co Ka
T y Bongigat US DAS al Vi 7 4 c U GEO
Z o DIO z DE X 4 d 8 yaon
ae Rî yc D Jf Palin Aj, [Sabino S a l d Y 2 S DS
Se to^ d LE و‎ Mie MN نافع‎ M 3 Lope ` SCH SS -
bd eri Í z ra J f uario oe ا‎ RAN o 9 y AE ری‎ Dauer
SI 3 Au N > Wl God RUNS A we ES j ; ya
$ 1 dali Jj 5 GË cal Date S «Fe ‘Sige T VW Chhunaparnd ^ ie A EC
x & Ke 2 © pal} Vë . 1 7 یں‎ 7:۷ NX E VEN Bora ROV 6ا‎ 5 1
3807 e imar) d / Chap: = ee WAF MAd enga? Bahari auke Maniario jj B jia \ hun Hot
Kë qe 7 ; 7 Kahaitdri Wea N a Ramdia ERGO iÑ OSA ا‎ 2 V. Të
a RL, V SH UE : 0 Wo art لے ہےر‎ ue Els N / o Kandali R
5 پش‎ of 0 کے‎ / diro ge 2 o 5 eT s D Jd 1 ET
E E ^ ER D ro ie A li SE SÉ -
2 È "i x 3 Uer eege =~ 763 «99^ D
atan, 3 o N plata
0 Grr,

(e
GE 2 i 7] doc, o ao e io
S EX En OSS ? s سے‎ ya p ES) A [an pees
duha. : N > 5ر‎ 7 LARGE Daly Se Wura Dalit
E P 1
dana JS

Lallipur © j y UE = Dluknaburi GE JE e $ ees A
nm 1 نتر‎ 2 Sc 7 f 2 AP رم یا ا‎

= k ; ی ےت‎ S AE wre : | e Be cT P 3 2

Vost SN 2 : 18 =

| COOCH |BEHAR

¡sar ] `
oa Kven

> Anal

Mair api

ZA omn

San

SMAI nook Bee

Sigs Balza SHED
SE
y ae

ود شر
و 343
P A‏
z :‏

ak Dë i .
Sra

JE

'SHILDÓNG
Rio pum E
Dags: y EN

3 06

E j ; Bret اف‎ : DC diy elle

1 dun 4 ze : EA : d ج2‎ ne SH =
Nn f i SN رج‎ 7

Í See . Sen : s m m

Shushdeg Duxgojur ? 3 ;

t کو‎ Hamgaen
j 7
4

9 ir
RR 400 A اس‎
pe zagli) ^ F. £ Kalin Em

3 ES = — 6ل‎ o a ef Ll a
5 =a DB "ad z x int

5
5

7۷۳۷۶
Wlan 2

Ashtodhar

Diaen
Netrak CG yo.
Netwakona yy

T P Gi
potas I ur d

o 5 e i :

7707

Haj. No 237, Dir, Geol. " 3۶+ 77

PART OF ASSAM TO ILLUSTRATE THE EXAMINATION OF THE EPICENTRAL TRACT OF THE GREAT EARTHQUAKE OF 12th JUNE 1897.

I
........... Route followed

® Pools and Lakes, not due to faulting
—— Faults and Fractures i

© Reported changes in aspect of Hills
Scale 1 Inch=16 Miles
(a ey 1 20 30 s0Miles

>.

E]

tm e Re m m
ور‎ d

Memoirs, Vol. X XIX, Map No. 3.

MAP
SHEWING BOUNDARIES OF THE

SEISMIC AREAS
oí the

AAA

118

During the 19th Century.

SeaiclInch 255 Mes :
) 106 soo Miles

Su-chawS

|
Ww

wa en

Y 7
| j ( eXumz
\ تحت‎ j |
Ni A "in kiang 4 ; |
D 0 Chin Mang, میڈ‎ |
PC [ing jung gI \ | à
1 >) o
zm) TERG ا‎ G E.
— ÓN
۰ y X green -kiang \ Tropic of Cancer. Bs.
2 8 - -- اپ ہے ہے == سح‎
Lage, کس سس‎

IGIPAL EARTHQUAKES OF

| CO nii den ` ES
کچ‎

7. D. Old

[gem

| ^ ہے‎
368
== |
i Y
ERN EG
am
B Su n
ہم‎
E ^
لے‎ 2
BA SY
| B4 Ya
H æ 5
06ے‎ $
2 RU 4 m
E
d

28°

IM - 0005

ts

Memoirs, Vol. XXIX, Map No. 3.
> = ei
100 00 109" |
— SSS ہے‎ ES
M A P

1
96°

GEOLOGIGAL SURVEY OF INDIA.
92

\

SHEWING ROUNDARIES OF THE

SEISMIC AREAS
of the

PRINCIPAL EARTHQUAKES OF
INDIA -

During the 19th Century.

Milos

rum

lebt
Dach Aj Vd
2 7 IN

Samber Sali.

: bi 1 ES

ic ian

5

cl A

mac,
Do

ng Luana
مد‎ raba

07 We 3
/ A S -
spon hg :
n مگ ے‎ 9

bad = pp
1

° pj

Muang ا‎

IR

Korali

Mg sen
و‎

Halle"! Ry aguda:
M

Cape Negros
Pagoda FÊ

‘Andaman H
* Barren L
J Ce

f
ا‎ 3

Sui

ES
$
E

8 80
A $
\ fp ما و‎ Ben
S
2. y stMathew A

[rae anam rt
Hoes Tande

000000
[Frere].

oF

Se
TOR
ا‎ U

2 b

Diaen

SE 3

5
7 2
dach"
i و‎

die TUN

et:
00 og 7

D
furrah

Fatal
o apy doo Trait

© Chinjan E
Tutt E
d
Dehra le
4 $

eu
Pda cilia

K

2-1۳20

ud‏ ا

A
,یی‎ Toa. Geh
E SE

e

Capo ملا‎

rater a

12

P

REFERENCES

MEL o

A

ce

و
]
1

- ^ -
- Lj
: pe aa " » K 5 à x e H
E P ^ 27 D D EZ Wi
H i - q + Ce? -— , i
WI . $ a $-
Kë H d? P * T -
^ ۔‎ - "AL i * e =
A ہیا‎ J d P ; ۰ 4 ا‎ > y E
3 D = d
à EI Ze? >. "
* D > " wä
D »- E *
a LJ E Lé L
D »
E el à 7 ۴ D
Le , [

D » 7 ۴
E B A Ke d
D * e D - B nd `
4 Y ` ‘ CR i A f 7 ~~
e > e d
E wë à z a E y S E AA 72 lo » 2 E 7
E , 4 F = Hy سے‎ a
یرت‎ - » ei سے8‎ 4 E D
A 3 مس‎ H d EF | "T
SE — A : t d E = Kë وم‎ Ke ^ q *
y e x ۹ PI. ا‎ Bt EET d 9 :
ae EN : K » D : We el T F ——— 000809
- E? ۶ P KR 2 N Fr T. ٠ a SA 5 » - ER
P ý گے‎ D e Wb 21 "Te Xe "= e 5 b à e
D H ^ Ee Aie RU L3 $
" e DW, rc 2 qe KE » Na f E EA
E ' e ے‎ " R
f 2 ot کاو‎ f E pe Va ls AE ہت‎ P 8
5 > 5 M و‎ E ELA ¡A e ER ai Saee: " wë?
d ne 92 = e سید‎ B = V | ` "ui M m D 2 > € A
: é ) Ki می‎ P E PUE, DT he d j í GEM y
d WE E d Wei D » a B
۱ p P» i E ww A ei d و‎ we Eë, Leer: 0 T D te e
AE 2 - P d y

Z
e ما‎

t
D
‘
£
f

H

(eg D d E Me :
۵ o 5 WË ds W و‎ paan a € E ۸ E
E. wr e£ db pM VM: i — i Pu T" mi 1
"e سن‎ Bt ee Oe EET D Seet | a . 7 مو ۔‎ E
0 `

D
Ye 2 - J 8 E E ` عم‎
T Ts l ‘ Re تب‎ n ھ1‎ mu ul رر‎ . d PE e 7۲ XT e = e
pea 2 i " i : سے *" 1 وی‎ A Pn . SÉ E رو ات‎ 0 cx 3 "EU i a ۵ A Cid a d a a Y i y ہے‎ E

"
i \
i à
x t
7 *
e oe & P
A
" d
`
« $
F ‘
G
‘ ] A 1
۰
1
5 H d `
a “
à
` J
S
E
e ~
A " ۱ à ۱
P LI
T `
i
i i>. i
. i d
, ` .
* L
í
5 ^
A
"
7 DH - f
à P
e
4 "
. ` D D
i *
^ L
Y
& i
A
Md D
1
[] D
P
۰
H e
۱ ,
j
p 5
"
۹ à
e +
7
1 d ۹ i
) : ;
/ 7
^ [4
e LJ
P
` i
à
y 5
D
a £ ; 5 *.
" , ,
D KA
‘ i ^
^ DE t 2
d
D ۸
F
i ` S
a E T 1
z 2 j 5
D J ‘ e - 8
: ’
Í
;
. i i ¿
b T
x ,
5
- D
j
A ;
D ^ P d
* US P
4 » »
T ۱
d b i و‎ ^ Tt
j K " [
y e x Ch
T^ la y ند‎
Y 1 I D
» D 4 D i ^
موس‎ t H T f

De