GENERAL ARTICLES

CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 35

Herbal medicine
V. P. Kamboj
Herbal medicines are the synthesis of therapeutic experiences of generations of practising physicians of
indigenous systems of medicine for over hundreds of years while nutraceuticals are nutritionally or
medicinally enhanced foods with health benefits of recent origin and marketed in developed countries.
The marketing of the former under the category of the latter is unethical. Herbal medicines are also in
great demand in the developed world for primary health care because of their efficacy, safety and
lesser side effects. They also offer therapeutics for age-related disorders like memory loss,
osteoporosis, immune disorders, etc. for which no modern medicine is available. India despite its rich
traditional knowledge, heritage of herbal medicines and large biodiversity has a dismal share of the
world market due to export of crude extracts and drugs. WHO too has not systematically evaluated
traditional medicines despite the fact that it is used for primary health care by about 80% of the world
population. However, in 1991 WHO developed guidelines for the assessment of herbal medicine.
Suggestions for herbal medicine standardization are outlined. The scenario and perceptions of herbal
medicine are discussed.
HERBAL medicine is still the mainstay of about 7580% of the
world population, mainly in the developing countries, for primary
health care because of better cultural acceptability, better
compatibility with the human body and lesser side effects.
However, the last few years have seen a major increase in their use
in the developed world. In Germany and France, many herbs and
herbal extracts are used as prescription drugs and their sales in the
countries of European Union were around $ 6 billion in 1991 and
may be over $ 20 billion now. In USA, herbal drugs are currently
sold in health food stores with a turnover of about $ 4 billion in
1996 which is anticipated to double by the turn of the century
1
. In
India, the herbal drug market is about $ one billion and the export
of plant-based crude drugs is around $ 80 million
2
. Herbal
medicines also find market as nutraceuticals (health foods) whose
current market is estimated at about $ 80250 billion in USA and
also in Europe
3
.
India is sitting on a gold mine of well-recorded and well-
practiced knowledge of traditional herbal medicine. But, unlike
China, India has not been able to capitalize on this herbal wealth
by promoting its use in the developed world despite their renewed
interest in herbal medicines. This can be achieved by judicious
product identification based on diseases found in the developed
world for which no medicine or only palliative therapy is
available; such herbal medicines will find speedy access into those
countries. Backward integration from market demands will pay
rich dividends. Strategically, India should enter through those
plant-based medicines which are already well accepted in Europe,
USA and Japan. Simultaneously, it should identify those herbs
(medicinal plants) which are time-tested and dispensed all over in
India.
The basic requirements for gaining entry into developed
countries include: (i) well-documented traditional use, (ii) single-
plant medicines, (iii) medicinal plants free from pesticides, heavy
metals, etc., (iv) standardization based on chemical and activity
profile, and (v) safety and stability. However, mode of action
studies in animals and efficacy in human will also be supportive.
Such scientifically generated data will project herbal medicine in a
proper perspective and help in sustained global market.
Herbal medicine
The World Health Organization (WHO) has recently defined
traditional medicine (including herbal drugs) as comprising
therapeutic practices that have been in existence, often for
hundreds of years, before the development and spread of modern
medicine and are still in use today
4
. Or say, traditional medicine is
the synthesis of therapeutic experience of generations of practising
physicians of indigenous systems of medicine. The traditional
preparations comprise medicinal plants, minerals, organic matter,
etc. Herbal drugs constitute only those traditional medicines which
primarily use medicinal plant preparations for therapy. The
earliest recorded evidence of their use in Indian, Chinese,
Egyptian, Greek, Roman and Syrian texts dates back to about
5000 years. The classical Indian texts include Rigveda, Atherveda,
Charak Samhita and Sushruta Samhita. The herbal
medicines/traditional medicaments have, therefore, been derived
from rich traditions of ancient civilizations and scientific heritage.
Nutraceuticals
V. P. Kamboj is in the Central Drug Research Institute, Lucknow
226 001, India.
e-mail: root@cscdri.ren.nic.in
GENERAL ARTICLES
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 36
This is a term of recent origin (1979) and comprises nutritionally
or medicinally enhanced foods with health benefits
3
. These include
engineered grain, cereals supplemented with vitamins or minerals
or genetically manipulated soybean and canola oil without trans
fatty acids, etc. Many pharma and biotech companies have moved
into this area since it does not involve regulatory clearances and
offers large markets. These companies have extended the term
nutraceutical to include pure compounds of natural origin like
lovastatin (a lipid lowering agent from red rice yeast),
docosahexaenoic acid (a cardiovascular stimulant from algae),
sterols, curcumin (from plants), etc. Likewise herbal preparations
are being marketed as nutraceuticals or health foods and even
the minimum standards laid down by WHO are not followed. It is
pertinent to mention that herbal medicines are therapeutics of the
indigenous/traditional systems of medicine and it is unethical to
classify them as health foods. The regulatory agencies should,
therefore, step in to prevent such misuse of natural
products/herbal medicines as was done by US-FDA by banning
the dietary supplement cholestin (i.e. lovastatin).
Nutraceuticals are in great demand in the developed world
particularly USA and Japan. Nutraceutical market in USA alone is
about $ 80250 billion, with a similar market size in Europe and
Japanese sales worth $ 1.5 billion
3
. Such huge markets have arisen
because of the Dietary Supplement Health Education Act passed
by USA in 1994 which permits unprecedented claims to be made
about food or the dietary supplements ability about health
benefits including prevention and treatment of diseases. This act
has motivated pharma to include not only compounds isolated
from fauna and flora but also herbal medicines as nutraceuticals,
which is unfortunate. The developing countries also see this as a
good opportunity and are marketing such products.
Herbal medicine market
As per available records, the herbal medicine market in 1991 in the
countries of the European Union was about $ 6 billion (may be
over $ 20 billion now), with Germany accounting for $ 3 billion,
France $ 1.6 billion and Italy $ 0.6 billion
3
. Incidentally in
Germany and France, herbal extracts are sold as prescription drugs
and are covered by national health insurance. In 1996, the US
herbal medicine market was about $ 4 billion and with the current
growth rate may be more than double by the turn of century. Thus
a reasonable guesstimate for current herbal medicine market
worldwide may be around $ 3060 billion. The Indian herbal drug
market is about $ one billion and the export of herbal crude
extracts is about $ 80 million (Table 1).
The 10 best-selling herbal medicines in developed countries
1
are
given in Table 2. The sales of these drugs account for almost 50%
of the herbal medicine market. These drugs have been well
standardized and some of them namely echinacea, garlic, gingko,
ginseng and saw palmeto are supported with mode of action and
clinical studies. Amongst the developed countries Germany holds
the lead and has published individual monographs on therapeutic
benefits of more than 300 herbs. In developing countries, China
has compiled/generated data on over 800 medicinal plants and
exports large amounts of herbal drugs. India has prepared only a
few monographs and its exports are dismal.
Why herbal medicine?
Herbal medicines are being used by about 80% of the world
population primarily in the developing countries for primary
health care. They have stood the test of time for their safety,
efficacy, cultural acceptability and lesser side effects. The
chemical constituents present in them are a part of the
physiological functions of living flora and hence they are believed
to have better compatibility with the human body. Ancient
literature also mentions herbal medicines for age-related diseases
namely memory loss, osteoporosis, diabetic wounds, immune and
liver disorders, etc. for which no modern medicine or only palli-
ative therapy is available. These drugs are made from renewable
resources of raw materials by ecofriendly processes and will bring
economic prosperity to the masses growing these raw materials.
Herbal medicine scenario in India
The turnover of herbal medicines in India as over-the-counter
products, ethical and classical formulations and home remedies of
Ayurveda, Unani and Siddha systems of medicine is about $ 1
billion with a meagre export of about $ 80 million. Psyllium seeds
and husk, castor oil and opium extract alone account for 60% of
the exports. 80% of the exports to developed countries are of
crude drugs and not finished formulations leading to low revenue
for the country. Thus the export of herbal medicines from India is
Table 1. Market size of herbal medicines
Country
Drug sales in
US $ (billion)
Europe (1991)
Germany 3.0
France 1.6
Italy 0.6
Others 0.8
Europe (1996) ~ 10.0
USA (1996) 4.0
India (1996) 1.0
Other countries (1996) 5.0
All countries (1998) ~ 30.060.0
Table 2. Ten best-selling herbal medicines in USA
Drug Botanical name
Market rank
as per sale
Echinacea Echinacea species 1
Garlic Allium sativum 2
Goldenseal Hydrastis canadensis 3
Ginseng Panax species 4
Ginko Ginko biloba 5
Saw palmeto Serenoa repens 6
Aloe gel Aloe barbadensis 7
Ephedra Ephedra species 8
Eleuthero Eleutherococcus senticosus 9
Cranberry Vaccinium macrocarpon 10
Table 3. Frequency of occurrence of medicinal plants in herbal
formulations in India
Common name Botanical name
No. of herbal
formulations
Triphala Terminalia chebula 219
Terminalia belerica
Emblica officinalis
Yashtimadhu Glycyrrhiza glabra 141
Pipali Piper longum 135
Vasaka Adhatoda vasica 110
Ashwagandha Withania somnifera 109
Mastak (Motha) Cyperus rotundus 102
Gulacha Tinospora cordifolia 88
Daruharidra Berberis aristata 65
Gokshura Tribulus terrestris 65
Kutaja Holarrhena antidysenterica 59
Punarnava Boerhavia diffusa 52
Source: BCIL
2
.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 37
negligible despite the fact that the country has a rich traditional
knowledge and heritage of herbal medicine. Considering the huge
herbal medicine and nutraceutical market in developed countries,
India should reconsider exporting crude herbal drugs.
Three of the 10 most widely selling herbal medicines in
developed countries, namely preparation of Allium sativum, Aloe
barbadensis and Panax species are available in India (Table 2).
India is the largest grower of Psyllium (Plantago ovata) and Senna
(Cassia senna) plants and one of the largest growers of Castor
(Ricinus communis) plant. These are also exported in large
amounts and yet our market share is dismal because of export of
crude extracts/drugs. Twenty other plants are commonly exported
as crude drugs worth $ 8 million. Five of these, namely
Glycyrrhiza glabra, Commiphora mukul, Plantago ovata, Aloe
barbadensis and Azadirachta indica are even used in modern
medicine. The plants Glycyrrhiza glabra, Piper longum, Adhatoda
vasica, Withania somnifera, Cyperus rotundus, Tinospora
cordifolia, Berberis aristata, Tribulus terristris, Holarrhena
antidysenterica and Boerhavia diffusa have been used in 52 to 141
herbal formulations and triphala (Terminalia chebula, Terminalia
belerica and Embelica officinalis) alone have been used in 219
formulations (Table 3). In spite of this, efforts have not been made
to preserve their germ-plasm from different localities,
identification of active plants vis--vis climatic zone and
development of agrotechnologies for their organized farming and
use as authentic materials in herbal medicines for better economic
gains.
India is one of the 12 mega biodiversity centres having over
45,000 plant species. Its diversity is unmatched due to the
presence of 16 different agroclimatic zones, 10 vegetative zones
and 15 biotic provinces. The country has 15,00018,000
flowering plants, 23,000 fungi, 2500 algae, 1600 lichens, 1800
bryophytes and 30 million micro-organisms
5
. India also has
equivalent to 3/4 of its land exclusive economic zone in the ocean
harbouring a large variety of flora and fauna, many of them with
therapeutic properties. About 1500 plants with medicinal uses are
mentioned in ancient texts and around 800 plants have been used
in traditional medicine; the most widely used plants are given in
Table 4. Tables 5 and 6 give the names of medicinal plants
exported and imported in India, respectively.
The major traditional sector pharmas, namely Himalaya, Zandu,
Dabur, Hamdard, Maharishi, etc. and modern sector pharmas,
namely Ranbaxy, Lupin, Allembic, etc. are standardizing their
herbal formulations by chromatography techniques like
TLC/HPLC finger printing, etc. There are about 7000 firms in the
small-scale sector manufacturing traditional medicines with or
without standardization. However, none of the pharma has
standardized herbal medicines using active compounds as markers
linked with confirmation of bioactivity of herbal drugs in
experimental animal models.
Table 4. Major Indian medicinal plants used in three indigenous
systems of medicine
Botanical name Sanskrit name
Abies webbiana Taleespatra
Achyranthes aspera Apamarga
Acorus calamus Vacha
Aloe sp. Kumari
Andrographis paniculata Bhoonimba (Kalmeg)
Asparagus adscendens Mushali
Asparagus racemosus Shatavari
Bauhinia variegata Kachnar
Bergenia ligulata Pashan bheda
Boerhavia diffusa Punarnava
Centella asiatica Mandukparni
Clerodendrum serratum Bharangi
Convolvulus pluricaulis Shankhapushpi
Crataeva nurvala Varuna
Dioscorea bulbifera Vidarikand
Embelia ribes Vidanga
Gymnemma sylvestre Madhunashni
Hedychium spicatum Shathi
Holarrhena antidysenterica Kutaja
Mesua ferrea Nagkesar
Nardostachys jatamansi Jatamansi
Ocimum sp. Tulsi
Phyllanthus amarus Bhumyamalika
Phyllanthus emblica Amalika (Amla)
Picrorhiza kurrooa Kutki
Piper longum Pippali
Pluchea lanceolata Rasna
Psoralea corylifolia Bakuchi
Rubia cordifolia Manjistha
Saraca indica Ashoka
Saussurea lappa Kushtha
Sida sp. Bala
Symplocos racemosa Lodhra
Terminalia arjuna Arjuna
Terminalia chebula Haritaki (Harad)
Tinospora cordifolia Guduchi
Tribulus terrestris Gokshura
Valeriana jatamansi Tagar
Vitex negundo Nirgundi
Withania somnifera Ashwagandha
Source: BCIL
2
.
Table 5. Medicinal plants being exported from India
Botanical name Part of the plant
Aconitum species
(other than heterophyllum)
Root
Acorus calamus Rhizome
Adhatoda vasica Whole plant
Berberis aristata Root
Cassia angustifolia Leaf and pod
Colchicum luteum Rhizome and seed
Hedychium spicatum Rhizome
Heracleum candicans Rhizome
Inula racemosa Rhizome
Juglans regia Bark
Juniperus communis Fruit
Juniperus macropoda Fruit
Picrorhiza kurrooa Root
Plantago ovata Seed and husk
Podophyllum emodi Rhizome
Punica granatum Flower, root and bark
Rauvolfia serpentina Root
Rheum emodi Rhizome
Saussurea lappa Rhizome
Swertia chirayita Whole plant
Valeriana jatamansi Rhizome
Zingiber officinale Rhizome
Source: BCIL
2
.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 38
Role of WHO in herbal medicine
Two decades ago, WHO referred to traditional health systems
(including herbal medicine) as holistic that of viewing man in
his totality within a wide ecological spectrum, and of emphasizing
the view that ill health or disease is brought about by an imbalance
or disequilibrium of man in his total ecological system and not
only by the causative agent and pathologenic evolution (WHO
6
),
probably implying that the indigenous system drugs (including
herbal medicine) restore the imbalance or disequilibrium leading to
the cure of ill health or disease. Such an attitude sent signals that
WHO as an organization has failed to provide leadership to
establish traditional systems of medicine which provide health
care to about 80% of the world population. However, it helped
the inclusion of proven traditional remedies in national drug
policies and regulatory approvals by developing countries. The
World Health Assembly continued the debate and adopted a
resolution (WHA 42.43) in 1989 that herbal medicine is of great
importance to the health of individuals and communities. The
redefined definition of traditional medicine thus issued in the early
nineties is given vide supra (see herbal medicine). Consequently, in
1991 WHO developed guidelines for the assessment of herbal
medicine
7
, and the same were ratified by the 6th International
Conference of Drug Regulatory Authorities held at Ottawa in the
same year. The salient features of WHO guidelines are: (i) Quality
assessment: Crude plant material; Plant preparation; Finished
product. (ii) Stability: Shelf life. (iii) Safety assessment:
Documentation of safety based on experience or/and; Toxicology
studies. (iv) Assessment of efficacy: Documented evidence of tra-
ditional use or/and; Activity determination (animals, human).
To the best of my knowledge, WHO has not systematically
evaluated any traditional medicine.
Herbal medicine standardization
In indigenous/traditional systems of medicine, the drugs are
primarily dispensed as water decoction or ethanolic extract. Fresh
plant parts, juice or crude powder are a rarity rather than a rule.
Thus medicinal plant parts should be authentic and free from
harmful materials like pesticides, heavy metals, microbial or
radioactive contamination, etc. The medicinal plant is subjected to
a single solvent extraction once or repeatedly, or water decoction
or as described in ancient texts. The extract should then be checked
for indicated biological activity in an experimental animal model(s).
The bioactive extract should be standardized on the basis of active
principle or major compound(s) along with fingerprints. The next
important step is stabilization of the bioactive extract with a mini-
mum shelf-life of over a year. The stabilized bioactive extract
should undergo regulatory or limited safety studies
Table 6. Medicinal plants being imported in India
Botanical name Native name
Cuscuta epithymum Aftimum vilaiyti
Glycyrrhiza glabra Mullathi
Lavendula stoechas Ustukhudus
Operculina turpethum Turbud
Pimpinella anisum Anise fruit
Smilax china Chobchini
Smilax ornata Ushba
Thymus vulgaris Hasha
Source: BCIL
2
.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 39
in animals. Determination of the probable mode of action will
explain the therapeutic profile. The safe and stable herbal
extract may be marketed if its therapeutic use is well
documented in indigenous systems of medicine, as also
viewed by WHO. A limited clinical tribal to establish its
therapeutic potential would promote clinical use. The herbal
medicines developed in this mode should be dispensed as
prescription drugs or even OTC products depending upon
disease consideration and under no circumstances as health
foods or nutraceuticals.
1. Rawls, R., C&EN, Washington, 23 September 1996, pp. 5360.
2. Sectoral Study on Indian Medicinal Plants Status, Perspective
and Strategy for Growth, Biotech Consortium India Ltd, New
Delhi, 1996.
3. Brower, V., Nat. Biotechnol., 1998, 16, 728731.
4. WHO, in Progress Report by the Director General, Document No.
A44/20, 22 March 1991, World Health Organization, Geneva,
1991.
5. Drugs and Pharmaceuticals Industry Highlights Incorporating
Patent Information, CDRI, Lucknow, 1998, vol. 21, pp. 3334.
6. Traditional Medicine, World Health Organization, Geneva, 1978.
7. Guidelines for the Assessment of Herbal Medicines, Document No.
WHO/TRM/91.4, World Health Organization, Geneva, 1991.
ACKNOWLEDGEMENT. I thank EMR (HRDG), CSIR, New Delhi
for the grant of Emeritus Scientist Scheme.
Received 30 August 1999; accepted 14 September 1999
The underground flower
Veenu Kaul*, A. K. Koul and M. C. Sharma
There are about 250,000 species of flowering plants in the world. Most of these produce flowers above
ground. Thirty six species bear flowers on underground shoots. Flowers produced above ground may
be of two types chasmogamous and cleistogamous. The former are the normal flowers which open to
receive pollen and/or pollinators. The latter type do not open and pollination is accomplished when
they are closed. Flowers borne on underground shoots are invariably cleistogamous. The
chasmogamous flowers are larger in size, produce copious amounts of pollen and large number of
small seeds. Contrarily, the cleistogamous flowers are generally reduced in size, produce little pollen
and few but heavier seeds
1
.
FLOWERING involves transformation of a foliar into a floral
bud through a series of histological, physiological and
biochemical changes
24
. Since flower represents a modified
shoot, and shoot is negatively geotropic, flowers almost
invariably differentiate above ground. If flowers had been
underground, the world would be devoid of the range of
colours, variety of scents, and innumerable patterns and
forms we see around us. The immense variety and
enormous beauty of flowers benefits the plants and appeals
the human eye. However, for the plant, underground flower
formation could be an asset, as it substantially cuts down
resource allocation involved in differentiation of accessory
floral parts, biosynthesis of pigments and production of
large quantities of pollen and nectar to reward pollinators.
Importantly, underground flowers have assured pollination
and seed set, with security against predators and vagaries
of environment. Nevertheless, the invariable differentiation
of flowers above ground has deprived plants of all the
above advantages. However, as compensation the above-
ground flowers confer on plants the ability for (i) cross-
pollination, which generates variability, assures adaptability
and evolutionary plasticity, and (ii) wider dispersal of pollen
and seed for greater distribution and reducing
intrapopulation competition.
That pollination and seed dispersal are the only major
events which aerial flowers help to accomplish is reflected
by Tulipa, Sternbergia, Ixilioron and such other bulbous
angiosperms in which flower development is completed
within the bulb, underground. The hidden flower is thrust
above ground for accomplishing pollination, whereafter,
seeds and fruits develop above ground.
Geocarpy
Have plants ever tried to combine the advantages of above
ground flowering and underground development of fruits?
The answer is provided by a few plants of which peanut
is the most common example. In this legume, flowers diffe-
The authors are at the Department of Botany, University of Jammu,
Jammu 180 006, India.
*For correspondence.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 40
rentiate above ground. Soon after pollination, they shed
their petals and bend with the help of a peg to first come
close to soil surface, and finally become subterranean. Pods
and seeds mature underground
5
. This phenomenon of
development of fruits underground is called geocarpy.
Besides Arachis hypogaea, development of aerial flowers
and subterranean fruits is also known in Trifolium
subterraneum, Voandzeia subterranea (L.) Thouars
6
and
Kerstingiella geocarpa Harms
6
.
There are also reports of the differentiation of fertile,
cleistogamous flowers on the underground shoots of pea-
nut
711
. Development of seed from flowers which are
ab initio underground, is called amphicarpy. However,
when these underground buds are brought above the
ground and exposed to light, they open readily, and follow
normal course of development
5
, suggesting that these
flowers are not truly cleistogamous and the plants are not
truly amphicarpic. Some scientists believe that such flowers
differentiate in response to the environment created by
farmers plow
5
.
Amphicarpy
During their evolutionary history, differentiation of true
underground flowers has been attempted by flowering
plants more than once. Of the nearly 250,000 flowering
plants, only 36 (Table 1; refs 1239) are amphicarpic. These
are distributed over 10 phylogenetically distantly related
Table 1. List of species that bear cleistogamous flowers on subterranean shoots
Family Valid name of the species Reference
Dicotyledons
Catanche lutea L. 1215 Asteraceae
(Compositae) Gymnarrhena micrantha Desf. 1417
Brassicaceae Cardamine chenopodifolia Pers. 18, 19; Fig. 1 a
(Cruciferae) Geococcus pusillus J. Drumm 14
Amphicarpa monoica (L.) Ell. 18
Vicia sativa ssp. amphicarpa (Dorth) 1215; Fig. 1 b
Fabaceae
(Leguminosae)
Aschers & Graebn.
Vigna minima (Roxb.) Ohwi & Ohashi 20
Lathyrus ciliolatus Sam. ex. Rech. f. 1315, 21
Pisum fulvum Sibth & Sm. var.
amphicarpum Warb & Eig. 1315, 22
Amphicarpaea bracteata (L.) Fern. 23, 24
Lathyrus amphicarpos L.
L. setifolius L. var. amphicarpos DC
Phaseolus sublobatus Roxb.
Tephrosia lupinifolia DC Prod.
Trifolium polymorphum Poir.
Galacita canescens (Scheele) Benth.
cf. 15
Polygalaceae Polygala polygama Walt. 18
P. pauciflora
Polygonaceae Emex spinosa (L) Campd. 15, 25
Polygonum thunbergii Sieb. et Zucc. 26
Scrophulariaceae Scrophularia arguta Soland. cf. 15
Urticaceae Fleurya podocarpa var. amphicarpa Engl. cf. 15
Violaceae Viola cucculata Ait., V. purpurea Kell., 18
V. sciaphila*
Monocotyledon
s
Commelinaceae Commelina virginica L. 18, 27
C. nudiflora L. 28, 29
C. indehiscens Barnes. 30
C. forskalaei Vahl. 18, 3134;
Fig. 1 c
C. benghalensis L. 18, 32, 33, 35;
Fig 1 d
Amphicarpum purshii Kunth 14, 15, 3639 Poaceae
(Gramineae) A. floridanum Chapman. 36
A. muhlenbergianum (Schult) Hitchc. cf. 15
Chloris chloridea (Presl.) Hitchc. 37
Eremetis (ca 4 species) cf. 15
Paspalum amphicarpum Ekman 36, 37
*Subterranean cleistogamy not confirmed.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 41
groups with a maximum concentration in the
Fabaceae (~ 10 species) and Poaceae (8 species). Most
amphicarpic plants are annuals; only a few are pere-
nnial. With a few notable exceptions, the amphicarpic taxa
grow well in aerated, well drained sandy or gravely soils.
The characters shared by most amphicarpic plants (see
Figure 1 ad) include the presence of (i) self-fertile sub-
terranean flowers that mature into large fruits and seeds
with limited dispersal, and (ii) aerial flowers that are capable
of cross-pollination and set many smaller fruits and seeds
suited to long distance dispersal.
Flowers of aboveground capitula of Gymnarrhena mic-
rantha Desf. a dwarf, annual desert composite, are chasmo-
gamous, while those comprising the subterranean capitula
are cleistogamous. The aerial capitula bear a large number of
small, wind dispersed fruits. On the contrary, the
subterranean fruits are large sized and fewer. They are never
shed; their seeds germinate in situ
16
.
Figure 1 ad. Hand drawings of plants of (a) Cardamine chenopodifolia; (b) Vicia sativa
amphicarpa; (c) Commelina forskalaei; and (d) C. benghalensis showing fruits/flowers (marked by
arrows) on aerial and underground shoots. r, roots; a, aerial; s, subterranean and d, diageotropic
shoots; af, fruits; afl, flowers on aerial shoots.
af
a
s
af
a
s
afl
a
d
s
r
r
r
afl
a
d
s
r
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 42
Aerial flowers of the plant are potentially open-pollinated,
which helps in increasing the genetic variability of the
population. The wind dispersed achenes widen the
distribution of the species to distant habitats. Subterranean
flowers are invariably self-pollinated and are therefore,
instrumental in preserving the parental genotype.
Underground fruits and seeds improve chances of survival
of these plants at specific microhabitats.
Amphicarpum purshii Kunth., an annual panicoid grass,
also bears aboveground and subterranean spikelets on the
same individual. The former are small and chasmogamous,
while the latter are large and cleistogamous. The
subterranean seeds are few but heavy. These account for
most progeny seedlings. Aerial florets outnumber the
subterranean flowers and contribute to the widening of
genetic variability of the species
38
.
Relative cost of aerial and underground flowers
as exemplified by Commelina sp.
In the genus Commelina of Commelinaceae, five species are
known to produce underground cleistogamous flowers.
Commelina forskalaei (Figure 1 c) and C. benghalensis
(Figure 1 d) bear flowers on three types of shoots
33,34
;
positively geotropic subterranean shoots, negatively geo-
Figure 2. a, An uprooted plant of C. benghalensis with exposed roots (brown) and
underground shoots (white) laden with flower containing spathes (white). b, Male (androgenic;
and.) and hermaphrodite (her.) flowers of an aerial spathe of C. benghalensis in bloom ( 3).
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 43
tropic cauline shoots and diageotropic shoots which run
parallel to soil surface. The positively geotropic leafless
shoots grow deep into the soil, and carry flowers inside
colourless spathes (Figure 2 a). Flowers on other two types
of shoot differentiate in green spathes and vary in number
as well as structure. In C. benghalensis spathes on the
diageotropic and subterranean shoots have a single
hermaphrodite flower each, but aerial spathes have three
and occasionally four flowers each. Flowers of aerial
spathes are trimorphic; the oldest is male and chasmo-
gamous, the second is hermaphrodite and chasmogamous
and the youngest is hermaphrodite and cleistogamous
(Figure 2 b). Flowers of the diageotropic spathes are always
chasmogamous, and those of the subterranean spathes are
invariably cleistogamous.
The subterranean cleistogamous flowers are obligately
self-pollinated. Their floral parts are small. The ratio
between the resources consumed in the differentiation of
their essential (stamens and carpels) and accessory organs
(sepals and petals) is 3 : 2 (60 : 40%). Resource expenditure
on pistil differentiation is 20% higher than that on the
differentiation of stamens. This is reflected in the greater
biomass of the pistil. Cleistogamous flowers have fewer
pollen grains and their pollenovule ratio is 2,305 : 1. They
also produce fewer but larger and heavier seeds than their
counterparts on aerial and diageotropic shoots. This
increase in size and weight of seeds is caused by the
diversion of resources saved from male function and
differentiation of extrafloral parts to the female function. The
diversion is made possible by assured pollination due to
cleistogamy despite the availability of fewer ovules borne
by the pistil.
On the contrary in all chasmogamous flowers, aerial as
well as diageotropic, greater share of resources is invested
in floral advertisement; it approaches 62% in male and 42
50% in hermaphrodite chasmogamous flowers. From the
total reproductive investment on chasmogamous herma-
phrodite flowers, 5661% investment is channelized to male
function. Even in cleistogamous flowers of the aerial
branches, the ratio between pistil and stamen biomass is
male biased unlike their subterranean counterparts.
Although anther dehiscence and stigma receptivity
overlap in hermaphrodite chasmogamous flowers leading to
self-pollination, these flowers hold the potential for cross-
pollination because of their colourful petals and anthers,
profuse pollen production and very frequent visitation by a
variety of hymenopteran insects. The potential for cross-
pollination can get expressed in the event of the failure of
self-pollination.
From what has been stated above, it follows that repro-
duction through underground flowers is less expensive, yet
more assured. In terms of investment/allocation of the
resources, the seed produced underground is cheaper
than that set above-ground. If selection pressure has not
worked against above ground flowers, and they continue to
differentiate on the plant, it is because they help in
retrieving one of the costs of sex, that of sharing gene(s)
40
through occasional outcrossing.
In most amphicarpic plants including C. benghalensis,
seedlings produced by subterranean seeds are more vigo-
rous than those produced by aerial seeds. As a con-
sequence, they have greater competitive ability and better
survival compared to seedlings resulting from aerial
flowers/fruits.
According to Cheplick and Quinn
14
, it is perhaps on
account of the importance of subterranean seeds to ind-
ividual fitness that they are produced early in ontogeny,
well in advance of aerial seeds. Zeide
17
has termed early
production of subterranean seeds and fruits a pessimistic
strategy of plants, suited to highly disturbed habitats,
where survival even up to the end of growing season is
uncertain. In such situations it is a definite advantage if
plants produce fruits as early as possible. In contrast, the
formation of aerial fruits is an optimistic strategy, whereby
reproduction is delayed until the end of growing season,
when time and growth conditions have resulted in
accumulation of sufficient resources in the plant body
15
.
Evolution of amphicarpy
What factors have led to the evolution of amphicarpy is a
question that remains to be answered. A number of hypo-
theses have been proposed from time to time. Since sub-
terranean seed production has evolved independently in
phylogenetically unrelated taxa, the factors underlying their
evolution are most likely to differ from species to species.
Mattatia
13,21
believes that amphicarpy in the genus Lathyrus
has arisen independently at least three times.
According to one hypothesis, the adaptive significance
of subterranean seeds is to expose them and the plants
differentiating therefrom to similar, presumably favourable,
microhabitat as that of the parent. However, in experiments
conducted on Amphicarpum purshii, plants raised from
subterranean seeds close to the parent did not always
outperform the plants raised at places far removed from the
parent
15
. A related possibility is that being better shielded
from the extreme fluctuations of microclimate at the soil
surface, the buried seeds retain viability, germinate and
establish seedlings far better than the seeds lying exposed
on soil surface. This hypothesis seems particularly
plausible for those amphicarpic plants which inhabit dry
habitats since active seed burial will, for instance, ensure
availability of greater soil moisture
15
. Supporting evidence
for this comes from Emex spinosa
25
, Amphicarpum purshii
15
and to a certain extent Commelina benghalensis. In all
these taxa, none (some in C. benghalensis) of the
subterranean seeds germinated when they were spread on
the soil surface. However, this hypothesis does not explain
evolution of subterranean seed production in species
inhabiting mesic environments.
Another hypothesis is that severe predator pressure
must have led to the evolution of subterranean seed pro-
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 44
duction which is understandable since buried flowers, fruits
and seeds are comparatively safe from foraging animals. In
grasses, it is easy to envisage the selective advantage of
subterranean seed production under conditions of intense
grazing
15
. Even in deserts where animals can be a major
cause of seed predation, seed burial might be an adaptive
response.
Detailed comparative data on the effect of herbivory
on aerial and subterranean flowers, fruits and seeds are
required to confirm this hypothesis.
Another advantage which can accrue to a plant from
subterranean seeds becomes explicit during a major distur-
bance which periodically destroys the aerial portion of a
herbaceous plant. At this critical juncture in plants life
cycle, only individuals producing subterranean propagules
would contribute to the formation of next generation. This
would be true especially for annuals which usually lack
vegetative propagation and therefore, have a single means
of reproduction. In Amphicarpum purshii
15
, Vigna minima
20
and Commelina virginica
27
, amphicarpy has been viewed
as possible adaptation to escape fire.
All the above hypotheses concede selective advantage
to subterranean reproduction. If this were so, why do
amphicarpic plants still produce aerial seeds? Is it for
combating the constraints associated with subterranean
reproduction? The retention of aerial flowers may be a
selective compromise between the risks associated with
production of either of the two types of flowers.
Detailed ecological, evolutionary and physiological
studies are required to fully appreciate the actual signifi-
cance of underground flowers.
1. Lord, E., Bot. Rev., 1981, 47, 421 450.
2. Esau, K., Plant Anatomy, Wiley Eastern, New Delhi, 1974,
2nd edn.
3. Cutter, E. G., Plant Anatomy Part 2: Organs, Edward Arnold,
1978.
4. Fahn, A., Plant Anatomy, Pergamon Press, Oxford, 1989, 3rd
edn.
5. Smith, B. W., Am. J. Bot., 1950, 37, 802815.
6. Tropical Legumes: Resources for the Future, National Academy
of Sciences, Washington, DC, 1981, pp. 47 53.
7. Meisnes, C. F., Lipsiae, 18361843, p. 97.
8. Kurtz, F., Sitz, 1875, 22, 4256.
9. Richter, C. G., Inaug. Diss kgt bot Gart Breslau A Schreiber
Breslau, 1899, p. 39.
10. Theune, E., Biol Pflanzen F Cohn, 1916, 13, 285346.
11. Bouffil, F., Sect. techd Agric. Tropicale Bull. Sci., 1947, 1,
1112.
12. Plitmann, U., Isr. J. Bot., 1973, 22, 178194.
13. Mattatia, J., Ph D thesis, Hebrew University, Israel, 1976.
14. Cheplick, G. P. and Quinn, J, A., Oecology, 1982, 52, 327
332.
15. Cheplick, G. P. and Quinn, J, A., Tree, 1987, 2, 97101.
16. Koller, D. and Roth, N., Am J. Bot., 1964, 51, 2635.
17. Zeide, B., Am. Nat., 1978, 112, 636639.
18. Uphof, J. C. Th., Bot Rev., 1938, 4, 2149.
19. Cheplick, G. P. and Quinn, J.A., Oecology, 1983, 57, 374379.
20. Gopinathan, M. C. and Babu, C. R., Pl. Syst. Evol., 1987, 156,
117126.
21. Mattatia, J., Bot. Notiser, 1977, 129, 437444.
22. Mattatia, J., Bot Notiser, 1977, 130, 2734.
23. Trapp. E. J., Am. J. Bot., 1988, 75, 15351539.
24. Trapp. E. J. and Hendrix, S. D., Oecology, 1988, 75, 285290.
25. Weiss, P. W., Oecology, 1980, 45, 244251.
26. Kawano, S., Hara, T., Hiratsuka, A., Matsuo, K. and Hirota, I.,
Plant Species Biol., 1990, 5, 97120.
27. Mendes, C. V. J., Beltrait, C. M. and Paoli, A. A. S., Rev. Bras.
Biol., 1994, 54, 403412.
28. Calvino, E., Mem. Soc. Cubana Hist. Nat., 1922, 4, 4577.
29. Calvino, E., Mem. Soc. Cubana Hist. Nat., 1923, 5, 99105.
30. Barnes, E., J. Bom. Nat. Hist. Soc., 1949, 46, 7089.
31. Hagerup, O., Dansk. Bot. Ark., 1932, 8, 120.
32. Trochain, J., Acad. Sci. Colon., 1932, 194, 743745.
33. Maheshwari, P. and Maheshwari, J. K., Phytomorphology, 1955,
5, 413422.
34. Maheshwari, S. C. and Singh, B., Phytomorphology, 1958, 8,
277298.
35. Maheshwari, P. and Singh, B., Curr. Sci., 1934, 3, 158160.
36. Weatherwax, P., Bull. Torrey Bot. Club, 1934, 61, 211215.
37. Connor, H. E., N. Z. J. Bot., 1979, 17, 547574.
38. Mc Namara, J. and Quinn, J. A., Am. J. Bot., 1977, 64, 1723.
39. Cheplick, G. P. and Quinn, J. A., J. Ecol., 1988, 76, 263273.
40. Schoen, D. J., Oecology, 1982, 53, 255257.
ACKNOWLEDGEMENTS. V.K. thanks the Head, Department of
Botany, University of Jammu for providing the necessary library and
laboratory facilities and to Council of Scientific and Industrial
Research, New Delhi, for the award of Junior and Senior Research
Fellowships.
Received 15 June 1999; revised accepted 28 October 1999
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 45
The Chamoli earthquake, Garhwal Himalaya:
Field observations and implications
for seismic hazard
Kusala Rajendran*, C. P. Rajendran, Sudhir K. Jain, C. V. R. Murty and Jaswant N. Arlekar
The Chamoli earthquake in the northern part of Uttar Pradesh is an important event from the point of
view of seismic hazard and risk assessment in the Himalaya. Tectonically, it is significant due to its
location in the central seismic gap, a 700-km-long segment between the 1905 Kangra (M 8.6) and the
1934 Bihar (M 8.4) earthquakes. Occurrence of two moderate earthquakes (1991 Uttarkashi and 1999
Chamoli) within a period of nine years naturally raises concern about the seismogenic potential of the
region. In this paper we present observations made during the post-earthquake survey around
Chamoli, and address some issues regarding the regional seismic hazard.
THE Chamoli earthquake of 29 March 1999 is yet another
moderate event of this decade in the Garhwal Himalaya.
It occurred at 00:35:13.59 h (local time) near the town
of Chamoli in northern India (Figure 1 a and b). The
US Geological Survey (USGS) located the event at
3049.2N, 7928.8E (m
b
6.3 and M
s
6.6), and the India
Meteorological Department (IMD) located it at 3017.82N,
7933.84E (m
b
6.8 and M
s
6.5; focal depth ~ 15 km). A long
aftershock sequence including at least three events of M > 5
followed the main event, some of which were located using
local and regional stations (Figure 2 a). The USGS fault-
plane solution indicates a pure thrust mechanism with two
nodal planes striking at 282 and 97 (Figure 2 b).
The earthquake triggered landslides, blocked several
roads, and disturbed electricity and water supply. A maxi-
mum intensity of VIII (MSK) has been attributed to this
event
1
. Maximum damages occurred in the district of
Chamoli where nearly 2600 houses collapsed and over
10,800 were partially damaged, leaving about 100 dead and
400 injured. The quake was also felt at far-off places such as
in Kanpur (440 km south-east), Shimla (220 km north-west)
and Delhi (280 km south-west). A few buildings in Delhi
sustained non-structural damages
1
.
The Chamoli event is important from various consi-
derations. One: its location in the central seismic gap
(Figure 1 a), a segment of the Himalaya that is considered to
have the maximum potential for a large earthquake
2,3
. Two:
its proximity to the high dam under construction near Tehri,
~ 125 km west of Chamoli. Here, we present some of the
observations made in the Chamoli area and discuss the
significance of this earthquake in our understanding of the
seismic hazard of the region.
Kusala Rajendran and C. P. Rajendran are in the Centre for Earth
Science Studies, P.B. No. 7250, Akkulam, Thiruvananthapuram
695 031, India; Sudhir K. Jain, C. V. R. Murty and Jaswant N. Arlekar
are in the Department of Civil Engineering, Indian Institute of
Technology, Kanpur 208 016, India.
*For correspondence. (e-mail: kusala@vsnl.com)
Figure 1. a. Sketch map of the Himalaya
11
showing the Himalayan
front (solid line). Meizoseismal area of four great earthquakes are
shaded in grey. Hatched area is enlarged in Figure 1 b; b. Simplified
geologic map of the north-western India
4
. Locations of the
Uttarkashi and Chamoli earthquakes are shown. The region where
maximum intensity was observed
18
during the 1803 earthquake is
indicated by solid circle. Location of Tehri dam is also shown.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 46
Geologic and tectonic setting
The Himalayan mountain range, an outcome of the com-
pressional processes ensued by the IndiaAsia collision
(7040 Ma) has been undergoing extensive crustal shor-
tening along the entire 2400-km-long northern edge of the
Indian plate. A series of major thrust planes the Main
Central Thrust (MCT), the Main Boundary Thrust (MBT)
and the Main Frontal Thrust (MFT) have been formed as
a result of these processes
4,5
. In some models, these thrust
faults are considered to have evolved progressively,
leaving the older ones dormant whereas in others, they are
treated as contemporaneous. For example, the evolutionary
model
6,7
considers the MCT to be an older thrust plane that
was more active in the early phases of the Himalayan
orogeny and MBT as a younger one that is more active
currently. The steady-state model on the other hand, treats
the MCT and the MBT to be contemporaneous and merging
at depths with a common detachment surface where the
great Himalayan earthquakes are believed to originate
8
.
The seismicity of the Himalaya, therefore, needs to be
understood in terms of the relative roles of these faults. It
has been argued on the basis of focal mechanisms
9
that the
MCT is probably aseismic and the current activity is on the
MBT. However, Chander
10
noted that the coseismic ground
elevation changes observed during the 1905 Kangra
earthquake could not be explained by assuming slip on the
MBT. The pronounced band of seismicity observed
beneath and south of MCT in Kumaun and Nepal
11,12
is
another indication of active deformation. The earthquakes
recorded during 19841986 by a network of stations in the
Yamuna and Bhagirathi valleys are also noted to be
following a trend of the MCT
13
. The 1991 Uttarkashi
earthquake (Figure 1 b) is the most recent activity
associated with the MCT
14
.
Tectonically, the MCT represents a ductile shear zone at
depth, comprising a duplex zone with three distinct thrust
planes: MCT I, MCT II and MCT III from south to north.
Based on the degree of metamorphism, lithostratigraphy
and tectonic setting, these thrust planes are also referred to
as Chail (MCT I, lower thrust), Jutogh (MCT II, middle
thrust) and Vaikrita (MCT III, upper thrust)
15
. Of these, the
Chail Thrust (MCT I), the southern-most and the youngest,
is believed to have moved during the Uttarkashi
earthquake
14
. The Chamoli earthquake appears to be
associated with the ongoing deformation along this thrust.
An active fold?
The Lesser Himalayan sequence lying between the MCT
and the MBT shows stacking of various groups of rocks
characterized by south-vergent imbricate thrusts, which
were later folded into major scale synforms and antiforms
15
.
Geological map of the area indicates presence of an
anticlinal structure very close to Chamoli
15
. The whole area,
considered as a schuppen zone, is delimited on
two sides by almost vertical faults the EW trending
Alaknanda fault in the south and the NNWSSE trending
Nandaprayag fault in the east
15
. Several parallel faults have
been mapped within this schuppen zone and one inter-
pretation is that, these faults demarcate isoclinal anticlines
split along the contacts of various litho-units
15
.
During the post-earthquake investigations, we observed
some signatures of recent deformation, associated with the
anticline mapped near Chamoli. A sharp contact of MCT I
with recent/sub-recent deposits was located on the
southern flanks of this anticline. Thick deposit of colluvium
(boulders and pebbles intercalated with coarse sand) occurs
at the foot of the steeper limb of the fold (Figure 3). The
colluvium may have been remobilized on an incipient slope
due to the development of the growing fold. Such surficial
features have been associated with fault propagation
folds
16
. We interpret the contact near Chamoli to be the
surface expression of an active fold. The tight
compressional folding in the Berinag quartzite and the
stretching lineation in mylonitic quartzite observed at these
localities are suggestive of the intense shortening along
this contact.
The above observations are significant because the
contact of the thrust plane occurs very close to the
Figure 2. a. Aftershocks until 8 April, as reported by IMD. Main
shock locations by IMD and USGS are also indicated; b. Fault plane
solution of the main shock (Source: USGS).
b
a
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 47
epicentral zone of the Chamoli earthquake. Although
the models for many earthquakes including the Uttarkashi
event suggest the rupture along MCT I
14
, geological
evidences for active faulting in this region are sparse. From
this point, the above observations from the epicentral
region of the Chamoli earthquake may provide certain clues
to identify active faults/folds in the Himalaya. The present
data by themselves are insufficient to suggest the nature of
the ongoing deformation in this region, but they provide
pointers for selecting sites for palaeoseismological and
related investigations.
Historic and current seismicity
Although four great earthquakes (M > 8) have occurred
along the Himalayan front during the last 100 years,
the Garhwal region is not known to have experienced a
magnitude 8 or larger earthquake in the recorded history
13
.
Historic and recent seismicity of the KumaunGarhwal
region (Figure 4) suggests the occurrence of at least three
earthquakes of M > 7 in this region. The largest historic
earthquake reported from this region occurred
on 1 September 1803 (ref. 17). Several villages were reported
to have been buried by rockfalls and land-
slides caused by this earthquake
18
. The Badrinath temple
located ~ 40 km north of Chamoli was severely damaged in
this earthquake. The epicentre based on the maximum
intensities is located ~ 100 km west of Chamoli
18
.
We examined two temples (7th12th century AD) at
Gopeshwar and Makkumath, both of which have been
reconstructed at least once in the past. Inscriptions on
stones, supported by historic data testify that the damages
to these temples caused by the 1803 event were substantial
and that the smaller structures around the main shrine were
totally destroyed. It should be noted that the temples at
Gopeshwar and Makkumath suffered only minor vertical
cracks during the 1999 earthquake, in spite of their locations
in the meizoseismal area, possibly because the 1803 event
was much larger. Based on the extent of affected areas, it
has been suggested that the 1803 event is a much larger
earthquake on the detachment surface
8
.
Intensity of shaking, site effects and coseismic
processes
The area affected by the Chamoli earthquake lies in seismic
zone V (IS:18931984), implying a potential for shaking
intensity of IX (and above) on the Modified Mercalli scale.
Our survey indicates that the maximum intensity of the 1999
event was only VIII (Figure 5). Intensity showed rather
abrupt changes from one location to another, probably due
to the local site conditions. For instance, the intensity of
shaking at Upper Birahi located on the river terrace was VIII,
whereas it was only VI at Lower Birahi located on the hard
rock (Figure 5). Similarly, Lower and Upper Chamoli showed
intensity VIII whereas Gopeshwar, located 2 km away on the
Figure 3. Contact zone of the Berinag quartzite (north-eastern
side) and the colluvium of pebbly sediments (south-western side)
developed on a growing anticlinal fold. This section is exposed on
the banks of the Alaknanda river near the Chamoli town. Whitish
rock (Berinag quartzite) in the foreground forms part of the uplifted
terrace (~ 1.5 m from the present river bed).
Figure 4. Historic and recent seismicity data (18031988) in the
Kumaun and Garhwal regions (Source: IMD, 1988).
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 48
hill slope, showed intensity V. Higher intensity observed at
Makkumath, located on the river terrace, ~ 15 km north-west
of Chamoli, is another example of site amplification.
Ground cracks developed at several places as part of
slope failure, causing threat to the settlements. Well-
developed ground cracks trending roughly in the eastwest
direction and showing lateral movement of up to ~ 12 cm
were observed at Gopeshwar, Chamoli and Bairagna (Figure
6). Attempts to make trenches across the ground fissures at
Telecom Hill in Gopeshwar were unsuccessful since these
were bottomed on the rubble and boulders at shallow
depths (~ 1 m), which form a part of the debris. In one of
these trenches, a poorly defined thrust plane was detected,
but its growth and overburden followed a complex pattern.
Although the trench sec-
tions did not reveal fault planes convincingly, the
fissures which had cut through concrete steps and well-
consolidated debris could be traced for nearly 1 km.
Orientation of these ground fissures, although disconti-
nuous, conforms to the trend of the MCT and also to one of
the nodal planes (282) inferred from the focal mechanism
(Figures 2 and 4). The predominance of eastwest oriented
fissures, particularly those developed in
the well-consolidated debris, may be manifestation of a
blind thrust.
The earthquake was also associated with marked changes
in groundwater discharge. In many groundwater springs,
flow increased by as much as ten times, surpassing even
the post-monsoon discharge. Flow decreased and the water
turned muddy, in one spring near village Bairagna, a
possible indication of fluidization and remobilization of fine
sediments.
Response of structures
The building stock in the affected area consists primarily of
rural dwellings, urban houses and a few modern
constructions. Load-bearing random rubble stone masonry
in mud mortar forms the predominant wall system. Brick or
concrete block masonry in cement mortar is used in many
newer constructions. The roofing system is usually thatch,
tin sheets, slate tiles, or reinforced concrete (RC) slabs.
Many recent constructions are in RC frames, with masonry
infill walls. In general, most of these are non-engineered
with no formal involvement of engineers in design or
construction. In this session we briefly discuss the
performance of common types of buildings in these areas.
Traditional stone dwellings
The traditional dwellings in the area are usually made up
of one or two storeys with a rather low storey height
(~ 1.65 m). The walls are about 0.450.60 m thick and are
made of random rubbles or slate wafers. The former type of
walls has two separate layers, the outer and inner wythes,
the intervening space being filled with stone rubble. In the
latter type, dressed stones and slate wafers are stacked
tightly using very little or no mud mortar. Most dwellings
have wood rafter roof supported directly on the walls.
Many old constructions and a few new buildings have
wood rafter roof supported on vertical wooden posts. Some
Figure 5. Intensity of shaking observed at various locations around
Chamoli. Shaded portion shows the trend of the fault as per the fault
plane solution which is consistent with the damage distribution. Figure 6. Ground fissure at Telecom Hill near Gopeshwar.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 49
of the new constructions use RC roof directly resting on the
walls.
Houses described above performed poorly, as expected,
and most deaths and injuries were caused by the collapse of
such constructions. Among these types of constructions,
those with masonry walls in slate wafers performed better
than those in random rubble masonry, probably due to
better interlocking in the latter. The most common damage
pattern was the separation of wythes following which the
walls tended to buckle (Figure 7).
Brick masonry buildings and buildings with
lintel bands
In general, buildings with burnt brick masonry in mud or
cement mortar performed much better than the traditional
stone masonry buildings. Numerous recent constructions in
stone as well as brick/concrete block masonry are provided
with a RC lintel band. Often rooms are provided with a RC
shelf of about half metre width, projecting from the wall at
the lintel level, serving the dual purpose of a storage slab
and a lintel band. Most houses with lintel bands performed
very well (Figure 8).
Reinforced concrete frame buildings
Many RC frame buildings (up to four storeys) with brick
masonry infill walls characterized by simple and regular
configuration, performed well even though most of these
were not formally designed, and certainly not for seismic
loads. The common form of damage included separation
cracks at the interface of the RC frame and infill panels, and
cracking of infill material.
Implications for the high dam at Tehri
Construction of the 260 m high rockfill dam at Tehri, located
between the MCT and the MBT (Figure 1) has remained
controversial since its inception. The environmental issues
associated with the dam as well as the seismic design
parameters have remained active topics of discussion
1924
.
Occurrence of another earthquake in its vicinity is likely to
enliven this debate. In this context, it may be useful to
review some of these issues.
Although no great earthquakes have been reported from
the vicinity of the dam during the historic past, the
Uttarkashi and the Chamoli events have occurred during a
span of nine years, within a radius of ~ 125 km from Tehri
(Figure 1 b). As mentioned earlier, the largest historic
earthquake in this region is the 1803 event of M > 7.
Maximum intensity based on historic reports
18
indicates that
the source of this earthquake may be within a distance of
50 km from the dam. Aside from current and historical
activity, this region is believed to have undergone several
movements in the recent geological past, as expressed by
the morphological features like deep incision of rivers and
development of river terraces
24
. The WNWESE trending
Srinagar Thrust is a prominent structure reported to be
passing through the vicinity of the dam
24
. Data on slip rates
or fault offsets in trenches are not available, placing major
limitations on the evaluation of recurrence rate of
Figure 7. Collapse of one of the wythes in a traditional house in
slate wafer masonry.
Figure 8. Two-storey house at Pipalkoti showing no damage. The
ground storey is in slate wafer masonry, upper storey in concrete
block masonry has been added later. Both storeys have RC lintel
band.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 50
earthquakes in this region. However, probability for an
earthquake during the projected life of the dam is
considered to be high
21,25
.
Effect of impoundment of a large reservoir leading to the
possibility of reservoir-induced seismicity (RIS) is another
concern. Proximity to an active thrust and geological
conditions favourable for infiltration of water into the deep
fault zones may favour weakening of faults, leading to
failure
20
. Gupta and Rajendran
26
suggested that the water
load might tend to stabilize the thrust faults
in the immediate vicinity of the Himalayan reservoirs,
making them less prone to seismicity, although the delayed
effect of pore pressure diffusion may be significant during
later periods. Mathematical simulation for the load-induced
changes at Tehri has also suggested a postponement of the
next earthquake
27
, but later studies suggest that the delay
may be short-lived
28
. Thus, the studies so far indicate that
water-induced weakening of the faults may remain as a
point of concern in the long-term life of the dam.
Another important issue is the possibility of landslips,
earthquake-induced or otherwise. A large chunk of land
falling into the river could generate large waves that could
breach the dam or could cause an overflow. Instances of
landslips that caused enormous floods in the Indus River in
Pakistan are reported
19
. The landslips and rockfalls that
followed a moderate earthquake at Chamoli (that too during
a dry season) underline the serious threat posed by these
processes and an urgent need to identify landslide prone
regions, from the point of seismic hazard associated with
high dams in the Himalaya.
Need for a database
A major issue of contention regarding the Tehri dam has
been the choice of peak ground acceleration (PGA). Pre-
liminary design of the dam was carried out by pseudo-static
analysis for a design seismic coefficient of 0.12 g.
Subsequently, dynamic analyses were carried out for
earthquake motions with effective peak ground acceleration
(EPGA) of 0.25 g, which was considered inadequate by
many workers
19,21,22
. Specialists from Russia have also been
involved in the evaluation of the seismic hazard at the dam
site and checking the dams safety. After considering a
number of postulated earthquake scenarios, their evaluation
of dam safety was based on two worst ground motions: a M
6.5 earthquake on Srinagar fault with PGA of 0.5 g at Tehri
site, and a M 8.0 event on the MBF with PGA of 0.4 g (ref.
29). At the time the dam was designed, strong motion
records were not available for this region, and the
characteristics of strong motion records obtained elsewhere
were used to develop the design spectrum. Data on stress
drop, attenuation characteristics and site amplification have
also been very limited, for a proper evaluation of the seismic
hazard
in the Tehri region. In this context, the earthquake at
Chamoli is significant as it provides a useful set of data.
Summary
The Chamoli earthquake gives further credence to the view
that the frontal parts of the MCT are still active and capable
of generating moderate/large earthquakes. From this point,
more studies in the epicentral region may be useful in
identifying and characterizing the active faults in the region.
Our studies indicate that it may be possible to identify sites
showing tell-tale signatures of active tectonism in these
areas, in order to quantify the slip rate. A moot point is
whether the faults in Chamoli and Uttarkashi are multiple
segments of a single structure, and if an earlier earthquake
occurred in 1803 and had ruptured both these segments,
resulting in a larger stress drop. If we assume this as the real
scenario, it is likely that the large and moderate earthquakes
in the GarhwalKumaun region follow a different rate of
recurrence.
Our studies suggest that the historic temples in the area,
which are among the oldest surviving structures, can be
used as archives of preserved evidences of past earth-
quakes in the region. A systematic study of these temples
may be useful to reconstruct part of the earthquake history
of these regions. The 1803 earthquake appears to be a larger
event (M > 7), which seems to have affected a wider area, in
comparison to the Uttarkashi and Chamoli earthquakes.
From an engineering point of view, damages due to the
Chamoli earthquake clearly demonstrated that codal pro-
visions for masonry houses are quite effective. While the
traditional stone houses failed as expected, constructions
with lintel bands performed well. Settlements developed on
alluvial terraces, although far separated, suffered severe
damages, while the intervening regions on relatively harder
rocks suffered little or no damage. Site-amplification could
be one of the probable reasons for wider damages during
the 1991 earthquake at Uttarkashi, which is situated on
extensive river terraces. The Chamoli earthquake provides a
whole set of new data, including new ground motion data to
study the seismic attenuation and site amplification
characteristics, for better hazard assessment and mitigation.
1. Jain, S. K., Murty, C. V. R., Arlekar, J., Rajendran, C. P.,
Rajendran, K. and Sinha, R., Special Report, Earthquake Engi-
neering Research Institute (EERI), California, 1999, vol. 33, pp.
18.
2. Khattri, K. N. and Tyagi, A. K, Tectonophysics, 1993, 96, 281
297.
3. Bilham, R., Curr. Sci., 1995, 69, 101127.
4. Gansser, A., Geology of the Himalayas, Interscience, New York,
1964, p. 289.
5. Molnar, P. and Chen, W. P., in Mountain Building Processes
(ed. Hsu, K. J.), Academic Press, New York, 1982.
6. Le Fort, P., Am. J. Sci., 1975, 275, 144.
7. Molnar, P., Chen, W. P., Fitch, T. J., Tapponier, P., Warsi,
W. E. K. and Wu, F. T., Editions C.N.R.S., 1977, 268, 269294.
8. Seeber, L. and Armbruster, J., in Earthquake Prediction: An
International Review, Maurice Ewing Series 4, AGU,
Washington, DC, 1981, pp. 259277.
GENERAL ARTICLES
CURRENT SCIENCE, VOL. 78, NO. 1, 10 JANUARY 2000 51
9. Ni, J. and Barazangi, M., J. Geophys. Res., 1984, 89, 1147
1163.
10. Chander, R., Tectonophysics, 1988, 149, 289298.
11. Yeats, R. S. and Thakkur, V. C., Curr. Sci., 1998, 74, 230233.
12. Pandey, M. R., Tandukar, R. P., Avouac, J. P., Lave, J. and
Massot, J. P., Geophys. Res. Lett., 1995, 751754.
13. Khattri, K. N., Chander, R., Gaur, V. K., Sarkar, I. C. and Kumar,
S., Proc. Indian Acad. Sci. (Earth Planet Sci.), 1989, 98, 91
109.
14. Cotton, F., Campillo, M., Deschamps, A. and Rastogi, B. K.,
Tectonophysics, 1996, 258, 3551.
15. Valdiya, K. S., Geology of Kumaun Lesser Himalaya, Wadia
Institute of Himalayan Geology, Dehra Dun, 1980, p. 291.
16. Powers, P. M., Lillie, R. J. and Yeats, R. S., Bull. Geol. Soc.
Am., 1998, 110, 10101027.
17. Ballore, M., Mem. Geol. Surv. India, Calcutta, 1904, 35, 153
194.
18. Smith, R. B., J. Asiatic Soc. Bengal, 1843, 12, 10291056.
19. Pearce, F., New Sci., 1991, 26, 3741.
20. Valdiya, K. S., Curr. Sci., 1992, 61, 801803.
21. Brune, J. N., Tectonophysics, 1993, 218, 281286.
22. Iyengar, R. N., Curr. Sci., 1993, 65, 384392.
23. Chander, R. and Kalpna, Curr. Sci., 1996, 70, 291299.
24. Valdiya, K. S., High Dams in the Himalaya, PAHAR, Nanital,
1993, p. 40.
25. Chander, R., J. Struct. Geol., 1992, 14, 621623.
26. Gupta, H. K. and Rajendran, K., Bull. Seismol. Soc. Am., 1986,
76, 205215.
27. Chander, R. and Kalpna, Curr. Sci., 1996, 70, 291299.
28. Chander, R., Kalpna and Gahalaut, V. K., Curr. Sci., 1996, 70,
873.
29. Thatte, C. D., Indian Geotech. J., 1992, 22, 141.
ACKNOWLEDGEMENTS. This study forms part of a DST
sponsored project at CESS, Thiruvananthapuram and another at IIT,
Kanpur, funded by the Ministry of Human Resources Development.
It was also partially funded by the Earthquake Engineering Research
Institute, USA. We thank IMD for the aftershock data, and the
district officials at Gopeshwar for support during the fieldwork.
Comments from three anonymous reviewers have been helpful in
revising the paper.
Received 14 June 1999; revised accepted 13 October 1999