Embed Size (px)
Citation preview
DEMOGRAPHIC CORRELATES OF URBAN DYNAMICS:
A CASE STUDY OF BANGLADESH
DISSERTATION SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS
FOR THE AWARD OF THE DEGREE OF
Maittx of ^ffHofiop^ IN
GEOGRAPHY
BY
REJUAN HOSSAIN BHUIYAN
Under the Supervision of Professor Abdul Aziz
DEPARTMENT OF GEOGRAPHY AU6ARH MUSLIM UNIVERSITY
ALIGARH (INDIA)
1988
DS1438
1 OS }Lj3e Ji!
ACKNOWLEDGEMENT
I acknowledge with deep respect my gratitude and express
sincere indebtedness to my supervisor and Departmental Chairman
Professor Abdul Aziz, for his ideas, suggestions and untiring
guidance throughout the M.Phil dissertation; his help and advice
rendered so patiently, is immeasurable.
I also feel highly indebtedness to Professor Mohammad Shafi,
for his useful suggestion. I also thank Professor Mehdi Raza,
Ex-Chairman, Department of Geography, for his encouragement.
Beside the Aligarh Muslim University, thanks are also due
to Professor Nazrul Islam (Departmental Chairman), Associate
Professor Rosie M. Ahsan and specially, Dr. Zia-us- Shams Haq,
Associate Professor, Department of Geography, University of Dhaka,
Bangladesh from whom I received many valuable suggestions.
I also greatful to Dr. Fakhuraddin, Lecturer of this
department and Mr.Mohammad Firoz Khan, for their valuable suggest
ion during the research period.
I owe much to Mrs. Shipra Banarjee, Mr. Mahmood A. Khan,
Mr. Shamim and others research scholar of the department, who
helped in various steps of the work.
Within the Bangladeshi friends, I am most obliged
to Mr. Kh. Mokaddem Hossain, Lecturer, Department of Sociology,
University of Dhaka and Mr. M. Asadur Rahman, Research Officer,
UNFPA- UNDP, Bangladesh, for providing the necessary secondary
data.
Both India and Bangladesh governments are duly acknowledged
for financial support in India which made this study possible.
I am greatful to my sick mother, two brothers and wife,
who also deserve special thanks for their moral support, for
bearance and instantaneous encouragement throughout my work.
I also offer my thanks to Mr. Najmuddin and Mrs. Rana Askari
Librarian, Seminar Library, for giving optimum facilities.
Finally, I thank Mr. H.S. Sharma for helping me in
electronic typing the M.Phil, dissertation.
(REJUAN HOSSAIN BHUIYAN)
CONTENTS Pages
Acknowledgemen t
List of Tables
List of Figures
List of Maps
CHAPTER - I :
CHAPTER - II
CHAPTER - III
INTRODUCTION
1.1 Key words and concept
1.2 Scope and objectives of study
1.3 Methodological consideration and Research Design
1.3.1 Study Area
1.3.2 Sources of Data
1.3.3 Data collection procedure
1.3.4 Data processing
1.3.5 Techniques of Analysis
1.3.5.1 Spatio-Temporal Distribution-R„ technique n ^
1.3.5.2 Hierarchy of Urban Centers: Composite index, correlation iratrix: and Rank size Rule
STATUS OF THE PREVIOUS RESEARCH
STUDY AREA-BANGLADESH
3.1 Historical setting
3.1.1 Origin, evolution and growth of urban system
3.1.2 Urbanization in the Newly Developing World (Third world countries)
3.2 Geographical setting
3.2.1 Location and Boundary
3.2.2 Area and Population
3.2.3 Physical setting
3.2.4 Identification of Region
CHAPTER - IV
CHAPTER - V
URBAN DYNAMISM: GEO-DEMOGRAPHIC AND SOCIO
ECONOMIC CHARACTERISTICS OF URBAN CENTERS ... 63
4.1 An over view of urbanization ... 63
4.1.1 Defination and concepts
4.1.2 Origin, e-.' olution and growth of
urban centers ... 68
4.1.3 Urbanization in Bangladesh ... 68
4.2 Nature and characteristics of
Urban growth dynamics ... 81
4.3 Interaction of the cities ... 96
4.4 Urban Population distribution and spatial pattern of urban centers.. 100
4.5 Spacing of Urban Centers of Bangladesh: Spatio-tempural analysis.. 113
4.6 Urban Hierarchy: Composite index and inter-variables relationship through correlation matrix . . . 125
4.7 Rank-size Rule Bangladesh ... 148
CONCLUSION ... 154
Appendices ... 158
Bibliography ... 167
LIST OF TABLES Pages
3.1
4.1
Percentage distribution of Population of 11 South Asian Countries by Urban-Rural
Intercensal growth rate of urban population, 1901-1981
43
75
4.2 Distributional Pattern of Urban Centers in Bangladesh (Number of Urban Centers by Urban Population Size)
4.2.1 Bangladesh Urban Agglomeration by Size group 1951-1981
4.2.2. Bangladesh:classification of districts by growth rate 1951-1961, 1961-1974, and 1974-1981
7 3
87
88
4.2.3 Selected Demographic, Socio-Economic Characteristics by Region for Bangadesh 93
4.4.1 Urban Population Distribution by Regions, 1951-1981 105
4.4.2 Urban Population as per cent of total popu-lation-1951, 1961, 1974 and 1981 by districts 107
4.4.3 Density of Urban Population by district, 1981 110
4.5.1 Nearest Neighbour Analysis 1951fa) 1961(b), 1974(c) and 1981(d)
4.6.1 Weighted Scores for the variables
4.6.2 BANGLADESH: Classification of urban centers on the basis of Composite Indices
122
132
146
4.6.2 11 X 11 Correlation matrix of 75 SMA and Municipalities of Bangladesh
4.7.1 Bangladesh: Rank-size Relationship, 1981
147
153
LIST OF FIGURES Pages
4.5.1 The Nearest neighbour, R^, Scale —
4.5.2 Measuring Nearest Neighbours ... 116
4.6.1 Scatter diagrams of Level of Hirrarchy, 15 ... 135 variables, 1988
4.6.2 Scatter diagrams of Level of Hierarchy, 4 ... 136 damographic variables, 1988
4.6.3 Scatter diagrams of Level of Hierarchy,11 ... 137 Social Variables
4.7.1 Rank-size rule of arithmetic scale ... 149
4.7.2 Rank-size rule of logarithmic Scale ... 149
4.7.3 Practicability of Rank-size relationship ... 151 Bangladesh, 1988
4.4.7 Percentage of Urban Population in the total in ... 106 each districts, 1974
4.4.8 Percentage of Urban Populatioa in the total in ... 106 each district, 1981
4.4.9 Densities of Urban Population, 1981 ... 112
4.5.1 Nearest Neighbour Distance, 1951 118
4.5.2 Nearest Neighbour Distance, 1961 ... 119
4.5.3 Nearest Neighbour Distance, 1974 ... 120
4.5.4 Nearest Neighbour Distance, 1981 ... 121
4.6.1 Spatial Distribution of First, Second and Third .. 138 order Functions, 4 Demographic variables, Bangladesh
4.6.2 Spatial Distribution of First, Second and Third ... 139 Order Function, 11 Social Variables,Bangladesh
4.6.3 Spatial Distribution of first, Second and Third ... 140 Order Functions, All 15 variables. Bangladesh
LIST OF MAPS Pages
3.1 Ancient City Map
3.2 Location of Bangladesh
3.3 Density of Population, 1988
3.4 Distribution of Population, 1988
3.5 Urban and Rural Popuation,1988
3.6 Rivers and Relief, 1988
3.7 Physiographic Divisions, 1988
3.8 Mean January Temperatures, 1988
3.9 Mean July Temperatures, 1988
3.10 Mean Annual Rainfall, 1988
3.11 Regional lUvision, 1988
4.1 Internal migration pattern, 1988
4.2.1 Growth of Urban Population, 1951-1961
4.2.2 Growth of Urban Population, 1961-1974
4.2.3 Growth of Urban Population, 1974-1981
4.3.1 Bangladesh: Urban Interaction Dynamics, 1988
4.4.1 Distributional Pattern of Urban Population Size, 1951
4.4.2 Distribution^]. Pattern of Urban Population Size, 1961
4.4.3 Distributional Pattern of Urban Population Size 1974
4.4.4 Distributional Pattern of Urban Population Size, 1981
4.4.5 Percentage of Urban Population in the total, in each districts, 1951
4.4.6 Percentage of Urban Population in. !:he total in each districts, 1961
APPENDICES Pages
1. Per cent and Rank of Urban Population by ...
district in census years 1961-1981 152
2. Distribution of Urban Population by ...
district in census years, 1961-1981 159
3. Bangladesh: Growth of Urban Population ...
1951-1961, 1961-1 )74 und 1974-1981 by
district 160
4. Ranking of Bangladesh Urban Centers (SMA ...
and Municipality) on the basis of
Hierarchical Concept (Demographic and
Social variables) 161
5. Ranking of Bangladesh Urban Centers (SMA ...
and Municipality) on the basis of
Hierarchical concept (All Variables) 164
CHAPTER - I
INTRODUCTION
Urban Dynamics are the certain cyclical proceses (Berg,
1987, p.2). It includes, some kind of kinetic process in which
growth and changes are prominant features, and determines the
technological, social, demographic, scientific, and political
platform. Moreover, some specilized urban factors that have
been changing the overall urban system. However, the urban
factors are a) Urban growth in relation to population, urban
growth in relation to function, urban growth in relation to
territory; b) Functional changes; c) Segregation of functional
areas; d) Invasion and succession; e) Trade off in functional
zones etc.
1.1 Key Wards and Concept:
Demographic correlates; Demographic correlates are infact
various demographic variables, like migration, urban population
growth rate. Non-agricultural population, fertility, mortality,
density of population, distribution of population, literacy
rate, size of the household etc.
URBAN DYNAMICS;
Urban Dynamics are the determinant of urban system change.
It is a kind of kinetic process in which growth and changes
are pronounced features. Urban systems are dynamic and changes
and this has an impact upon the system itself and its environ
ment. Moreover, this growth and change alters the spatial
structure and relationships of the entire system. The question
that arises is what are the urban dynamics determinants? In
this study following determinants of urban dynamics are used.
1. 1. Urban growth in relation to population: In 1901, the
urban population was only 7 lakh (2.43 per cent) but in 1951
this figure rose upto 18 lakh (4.33 per cent) and in 1981 census
it was 135 lakh (15.54 per cent). Urban dynamics factors are
the responsible for high rate of population growth in urban
areas of Bangladesh. Shifts in urban size categories exhibit
an important pattern of urban growth. Data on changing distri
bution of urban places by population size represent only the
net changes. But evalujiing the actual shifts in size categories
for particular places over the years permits a more accurate
evaluation of urban development.
2. Urban growth in relation to function; i.e., the functional
change which has made towns and cities distinct from other
settlements, more specially, the dominant function of towns
in order to find out to what extent they show functional speci
alization. Urban functions are predominantly those which depend
upon accessibility, whether mutual proximity or tansportation.
The following can be regarded as characteristically urban func-
tions-
a) Central place functions, or general services, which are
carried out for a more or less extensive but contiguous
area ;
b) Transport functions, which are carried out along the major
lines of communications,
c) Specialized functions, which are carried out for non-local,
non-contiguous areas. These could include extrative and
manufacturing industries with world-wide market, or indeed
minor industries whose distribution areas are smaller than
the general service area.
3. Urban growth in relation to territory: During the time
of evolution, an urban center might be very small in areal
extent, but over the time period, that area of the urban
center can increase in population.
II. Functional changes: The initial function or functions
often change with the passage of time, old functions decline
and new functions emerge as a result of which a city's
initial internal characteristics also change.
So an urban center can change it's character over
the period of time cities classified as manufacturing
(industrial). Commercial, retailing and diversified are
the most numerous. Other groups wholesaling, transportation,
mining, educational, resort or retirement, and others inclu
ding university and political factors are affecting urban
growth.
III. Segregation of functional Areas; After a optimum period
of time a urban center became a highly specialized and func
tional segregation starts through areal grouping; like high
class, middle class or lower class residential areas. It
means a particular type of functional activity segregates itself
in a specific area or zone. Sometimes racial, social, cultural,
economic and religious pregudices are also factors in the segre
gation of population.
IV. Invasion and succession; Due to the impact of urban dynamics
determinents, an area with respect to time series - gradually
change her character from one specific indentity to others
identity. It can be sited that Botanists have found that commu
nities of original species undergo a process of displacement
by different species. In human society the process is much
the same. Invasion occurs when a group of inferior economic
or cultural status moves into an area occupied by a group
possessing superior status, gradually taking over the district
and changing its complexion to match the culture of the invading
element. However, the process does not always take the form
of a displacement of a "superior" by an "inferior" group; oft-
times the reverse is true.
Factors initiating invasion : Cressey mentions ten conditions
which be believes may be associated with the invasion process:
"1) desire for increased social prestige, 2) pressure of wife
and children, 3) increased economic resources, 4) desire for
better living conditions, 5) activity of real estate agents,
6) desire for home ownership, 7) pressure of vacant homes,
8) changes in transportation facilities, 9) desire to be near
one's place of employment, 10) movement of industrial area.
And some other person also sited the reasion for invasion,
like Gibbard (Cressey, 1930, p. 227).
On the other hand succession is part of the invasion
process- the final stage or climax. The concept refers to the
stage of invasion in which displacement is complete or nearly
so- the stage in which a different population group or set
of institutions tends to predominate.
Mosque replace Mondir; or Mondir replace Mosque, depending
on which group is the invader and which the invaded. So, all
of these factors are the examples of 'Urban Dynamics'- which
are change the nature and characteristics of a respective city-
area.
V. Trade off in functional zones; It is one kind of relocation
process of different function of urban systems. It means one
kind of trade practice or business can shrink or close their
activities and relocate for another place.
Moreover, the behavioural feature is one of the urban
dynamics characteristics, here the change comes in fashion,
manner, conversation, and behaviour pattern of the inhabi
tant over time.
1.2 Scope and objectives of the study:
Cities are not static things. They grow in time and space.
Cities put together make the urban system. This urban system
like cities is not static it is dynamic and subject to change.
The scope of the study is to investigate the demographic deter-
minents of urban dynamics for 75 Statistical Metropolitan and
Municipal towns and cities of Bangladesh. Urban dynamics is
the outcome of many factors-political, economic, social, demog
raphic and scientific and technological. The purpose of this
study is to single out the role of demographic factors and
to assess its contribution to urban dynamics. The time for
which demographic correlates are studied is from 1951 to 1981,
a period of 30 years. It is during this period that external
and internal factors are impacting the change to all inter
and intra urban dynamism. An urban settlement is a dynamic
place where inter linked changes generated by certain cyclical
process occur in their internal structure. These changes them
selves are in a cyclical pattern and thus a several stages
can be distinguished. If it is so then there is the need to
develop an integral theoretical concept of urban dynamics and
to investigate the nature of urban systems and to investigate
the kinetic force of urban dynamism.
The primary objective of the study is to describe and
evaluate the general pattern of urban system change and identify
the causes and consequences of urban dynamics machanism, through
several statistical technique (for detail see methodology).
The applicability of urban system change determinant as a case
study of urban centers of Bangladesh has been attempted. The
study moreover, will focus, urban dynamics in relation to
different demographic, socio-economic variables. Urban popula
tion growth, functions of urban growth and territorial growth
of urban centers of Bangladesh, has been made in depth. Urban
dynamics machanism has been examined through hierarchical index.
Correlation matrix, gravity model, rank-size rule and specially
spatio-tempural distribution of urban centres from 1951 to
1981 through Rn - technique. The specific objectives of the
study are:
i) To present an overview of the urbanization of Bangladesh
historically;
ii) To know the origin, evolution and growth of urban
system of Bangladesh;
iii) To compare the situation of urbanization with relation
to developing countries;
iv) To see the different views of urbanization over the
world;
v) To study concept of urban, urbanism, urbanize, and
urbanization;
vi) To review the literature of urban dynamics with rela
tion to demographic correlates;
vii) To know the characteristics and factors governing
urbanization from the urban dynamics point of view;
v-iii) To examine the nature and characteristics of urban
centers of Bangladesh;
ix) To identify the strong veriables of urban dynamics
through correlation matrix;
x) To examine the Interaction of the cities;
xi) To observe the urban population distribution and
densities and specially, spatial pattern of urban
centers.
xii) To evaluate the spacing of urban centers, through-
Nearest Neighour analysis (Rn-technique) from 1951
to 1981;
xiii) To know the hierarchical situation of urban centers
through - Demographic and socio-economic variables
to use the composite index for the measurement of
urban dynamics determinants; and
xiv) To examine the rank size rule's applicability in
case of Bangladesh situation.
1.3 Methodological Consideration and Research Design:
In the case of Bangladesh the theoretical and empirical study
of urban dynamics arguments have been presented in this inves
tigation. The author has also tried to explore the machanism
of urban system change based on the principle that urban dynamics
is primarily the result of the spatial behaviour of urban factors
(like-Demographic, socio-economic, political,cultural, religious,
etc.).
1-3.1 Study Area: Bangladesh - stretches latitudinally between
20°30'N and 26M5'N and longitudinally between 88°E and 92°56'E.
It is one of the most crowded rural areas in the world. With
100 million people (1981) in 143,998 Sq. Km (55,598 Sq. miles).
1.3-2 Source of Data: Related data and literature was collected
from various related research papers both at home and abroad.
Most of the data is from government sources. Demographic and
socio-economic data have been collected mainly from district
census reports of 21 volumes, report on urban area of Bangladesh,
year book of Bangladesh by the Bangladesh Bureau of statistics
(BBS), 1981 to 1987 reports, and Bangladesh Pourashava
(Municipality) statistics, 1987, National Institute of Local
Government (NILG). Migration data have been collected through
pretested standarized questionaire.
1.3.3 Data Collection Procedure; All secondary and tertiary
data have been gathered from BBS, NILG and varieus research
organization like, center for urban studies (CUS), Dhaka
University, Bangladesh and all tertiary data from research section
seminar and Moulana Azad Library, AMU, India. Primary data
from field investigation in 4 SMA and 71 municipalities.
1.3.4 Data Processing; All data have been processed by manual
technique, including Rn-statistics, Hierarchical level, 11 x 11
correlation matrix and Rank-size rule etc.
1.3.5 Techniques of Analysis; Various statistical techniques
have been used to under stand the urban dynamics machanism of
Bangladesh. Gravity model for interaction of the cities, spatio-
temporal analysis (Rn-Technique), urban hierarchy by composite
index, inter-variables relationship through correlation matrix.
Rank-size rule etc. have applied for understanding the urban
dynamics of urban centers of Bangladesh.
1.3.5.1 Spatio-Temporal Distribution Through Rn-Statistics;
Last 4 decades of urban development in numbers has been
used as a measurement period. The Nearest-neighbour analysis
as outlined by Clark and Evans and as exemplified by king has
been applied.
1.3.5.2 Hierarchy of urban centers-through composite index
correlation matrix and Rank-size rule; Through corre
lation matrix, it has been tried to observe the inter-correlation
amongs the 11 x 11 variables. Those which are highly correlated
have been identified for composite index development. Three
different levels of hierarchy have been formulated.
According to Zipf's formulated Rank-size Rule the urban
centers progression is Pn = P.. R~ or 1, 1/2, 1/3, 1/4, 1/5...
1/nth rank but in case of Bangladesh, the top 13 cities did
not fit into the said formula. But in practical field it has
been found as a reverse 'J' slope.
CHAPTER - II
STATUS OF THE PREVIOUS RESEARCH
'URBAN DYNAMICS' is one of the most challenging field
in urban geography study. The scope of this chapter is (i) to
review the work done by the geographers and other associate
researchers in the field of population and urban geography
(ii) to examine the trends of research, and (iii) to identify
significant areas of research to be developed in the years
ahead.
A large number of literature on urban geography, parti
cularly on urban dynamics and changing pattern of settlements
of the developed countries has grown during the present century,
but our knowledge of the current dynamic processes,mechanism,
configurations and implications of development of urban dynamics
and urban system study in the developing countries are still
very limited. However, in Bangladesh both macro and micro level
urban dynamics studies are not rare but few and far between
in urban geographic research field (Rejuan, 1984).
There are at present two broad branches of geographic
research which have a bearing upon urban dynamics i.e. urban
geography and population geography. These are directly and
indirectly related to urban dynamics research.
2.1 Population geographer's and Demographer's views:
Systematic population study is one of the most recent
and youngest branches of geography (Gosal, 1972). A number
of population studies had been published before Trewartha
(Trewartha, 1953) but most of those studies were related to
distribution, density and growth, and they almost completely
neglected other dynamic characteristics. Recently a significant
11
growth of literature in population geography has taken place.
This is evident from Zelinsky's Bibliographic guide' and sub
sequent books and articles.
In demography, which also studies population but the
focus is on measuring or determining the rates, ratios and
trends of different population characteristics. Amongst the
earliest work in population dynamics and socio-economic charac
teristics with relation to urban system change of the world
perspective is the work of D.J. Bogue," who did a comprehensive
and pioneering work.through mathematical model of population
growth (Bogue, 1969). Nevertheless, some tangential and speradic
attempts of Locus (1979), Demko (1970), Meadows (1972), Brown
(1974), Woods (1979 and 1982) and Clarke (1981) were there,
but were not directly related to urban expansion factors and
population growth. These authors have however used the urban
growth variables in different phases of their studies. Before
them T.R. Malthus (1966-1834). M.T. Sadler (1780-1836), T.
Doubedly (1849-1902), and D. Ricardo (1772-1823), postulated
that there exists an inverse correlation between the increase
in population and the increase in food supply.
A good deal of work has been pursued on urban population
distribution and density in population geography by Akindobe
(1977), Tusi and Bogue (1978). They have examined the population
explosion in Africa and its implications for economic develop
ment. Akindobe discussed four main problems associated with
the African population explosion namely (1) percentage of
dependents, (2) growing urban unemployment, (3) rural unemploy
ment, and (4) percentage/number of skilled workers. All these
things, he tried to correlate these with urban growth phenomena.
Tusi and Bogue have examined the percentage of urban population
and its socio-economic composition. The authors have also
projected the world urban population and its danger level.
12
Gibbs (1961) has mentioned that the word urban generally
connotes demographic attributes (size and/or density) or
economic variables (the prevalence of non-agricultural occupa
tion). He also discussed the different terms of urban phenomena.
Like "Urban", "urbanism" and "urbanize". Petersen (1961), Bose
(1930) Clarke (1981) and Mehta (1978) have discussed the impact
of demographic correlates on urbanism. Petersen tried to define
'urbanism'. Day (1978) explored the implications of a population
growth with regard to health, economic conditions, education,
the place of women, social security, etc. as a urban dynamics
factors. The author also tried to assess the non-demographic
factors such as absolute size of the population which affects
the urban social conditions more than the age-sex structure.
A paper entitled 'Inida: the population Explosion and
Demographic Change' was published by Domros (1984). He has
examined the dynamic variables of urban growth mechanism.
Krishan and Bala (1982) examined the impact of the inter
national border on the growth of the population of adjoining
towns. They hypothesised that towns are likely to suffer in
their growth due to border situation. The data of eighty two
towns of India, adjoining the Pakistan border was analysed,
which point out that the proximity of border would discourage
new investments of industry, trade and even house construc
tion in frontier zone, more especially when the relations between
the two nations are not good. Due to that Rural people migrate
in nearest urban centre. Cities do not grow up themselves
(Jefferson, 1931), it has been argued that cities grow mostly
by net in-migration, and it has also been argued that they
grow mostly by their own natural increase (Keyfitz, 1980).
Kingsley Davis finds that cities in the European industrial
revolution grew mostly by in-migration but cities of the contem
porary less-developed countries are growing mostly by their
13
own natural increase (Davis, 1965 and Rogers, 1977). Moreover,
the case for their growth by migration was expressed as far
back as Slissmilch, who showed that deaths exceeded births in
the principal cities of Europe, so that without in-migrats
cities would decline. The same view has been developed by Fischer
(1976) and Wrigley (1969). Most recently Sharlin (1979) found
the contrary for the cities of Europe. He collected evidence
to show that the population native to the city did replace
itself, and that the excess of deaths over births that appeared
in the city did replace itself, and that the excess of death
over births that appeared in the overall statistics was due
to temporary residents, especially tradesmen and servants who
had no chance to marry but whose deaths would be counted as
urban if they died in the city (Keyfitz, 1972). Keyfitz (1980)
also tried to explore the impact of female births on urban
population growth.
For the urban system change, the demographic variables
play a very important role. Age structure, sex compostion,
marriage, fertility, mortality, migration, etc. contribute
a great deal to changing the urban atmosphere of a country.
Social and economic set up, also contribute to change in the
urban structure and environment as a whole. These are the housing
status, education system, income level, religious attitude,
attitudes towards family size and other administrative variables.
However all of these variables influence the degree of urban
concentration, spatial extention and structural change of urban
atmosphere (Chandra, 1986). He emphasised that fertility, morta
lity and mobility are the three basic components of population
change and the factors affecting these basic components are
the real determinants of population change in urban sectors.
Innumerable studies have been done by Indian population
and demography researcher on fertility in urban communities.
The studies of Chatterji (1967) are most exhaustive. He inves-
14
tigated the fertility levels and population distribution of
urban population in the country's seven macro-regions. He also
examined the role of physical & historical factors, the past
history of India as a human habitat, and politial events in
introducing regional variations in populations distribution
and their impact on urban ecology. There is considerable litera
tures concerning fertility and urban growth; like that of
Federici (1968), Smita (1983), United Nations (1953) etc.
The causes of mortality vary both in time and space.
Spatially, different urban areas are at different status of
socio-economic development and technological advancement. However,
a broad distinction has been made between endogenetic (biological)
and exogenetic (environmental) factors by Chandra (1985). He
also be classified determinants of mortality into three basic
categories demographic, social and economic. The same view
has been expressed by Mehta (1978), Brass,(1971) , Rama Kumar
(1986), Luces (1979),Preston (1975).
Equally, age-structure of a population has been mentioned
as the most prominent demographic factor governing the incidence
of mortality and level of fertility in an urban community.
Sex composition is another basic demographic determinant of
fertility. Urban centres mostly attract male in-migrants and
it leads to the factor of residence or the degree of urbani
zation (Chandra, 1986).
Like fertility and mortality, migration is one of the
dynamic constituents of population change in any urban area
(Trewartha, 1969).
Kayfitz (1980) has studied the contribution of the two
components of city growth under the various hypothetical
conditions. The conditions were (i) net in-migration and (ii)
Urban grew mostly by its own natural increase. He also tried
15
to check the statement through mathematical derivation. He rt starts from the simple notation (P = P e ) to very complex
calculation. If there is no city population there can be no
natural increase, and during the time after a city is established
but still small, its births cannot be numerous. In the other
extreme, when the country is mostly urbanized there is little
rural population left to migrate to cities.
In this connection out of three major components of
population change, migration is the most difficult to concep
tualise and measure (Jones, 1981). The difficulties in concep
tualising and measuring the phenomenon of rural-urban migration
arise because, unlike fertility and mortality, the rural-urban
migration is not just an unequivocal biological event but a
physical and social transaction also (Zelinsky, 1971). No wonder,
compared to other attributes of population, migration has
attracted the maximum attention. Lee (1969), Prothero (1979),
Gosal (1961), Bogue (1959), Garnier (1966), Smith (1960), Chandra
and Sidhu (1980), Gosal and Krishan (1975), Mehta (1978), Chandra
(1986) and Hartshorn (1982) have discussed causes of rural
to urban migration that carries the rural folk to the growing
urban centers a feature which is more pronounced in the less
developed countries. It is caused by both "push" and "poll"
factors. In the less developed world which has high rural den
sities and where some urban industrial development is taking
places, both push of the rural areas and pull of the urban
areas generate migratory tendencies among the people. In rural
areas appalling poverty, unemployment, low and uncertain wages,
uneconomic land holding tenureal system and poor facilities
for education, health, recreation and other functional services
work as the push factors. By comparison, the pull of the urban
areas may include better employment opportunities, regular
and high wages, fixed working hours, better amenities of living,
facilities for education and socio-cultural activities. Moreover,
16
the life in urban areas tends to be more attractive and secure,
though there has been recent spurt in urban vice both in deve
loped and underdeveloped worlds. In countries like where more
acute poverty and natural hizards in village areas. As a
result of these factors rural to urban movement takes place.
However, in Bangladesh where the big cities are the recipients
of rural migrants on a large scale (Islam 1976), large number
of slums emerge in such cities. The big cities in Bangladesh
(like, Dhaka, Chittagong, Khalna, etc.) are usually industrial
and port cities and get their labour from the vast reservoir
of surplus rural manpower. He also found that such cities are
unable to provide even the minimum amenities to these migrants
and it has given rise to slum conditions in ports cities.
2.2 VIEW OF OTHER URBAN RESEARCHERS
Having reviewed the work of population geographers and
demographers it will be in fitness of things that the works
of other urban researches are also reviewed. For understanding
URBAN DYNAMICS, it would be necessary to identify the nature
and characteristics of urban system as a whole and to determine
the correlates of this system and also investigate the
a) Spatio-Temporal distribution of urban centers,
b) Hierarchy of urban centers, and
c) Rural-urban linkages
Although, urban geography is the geographic study of
urban places which evolve, grow and exist as service centres,
largely for their surrounding areas (Gosal, 1972). Urbanizaion
is the aggregation of people into large, dense and heterogeneous
settlements (Thakur, 1980). Urban Dynamics are the determinents
of urban system change. It is a kind of kinetic process in
which growth and changes are prominent features. Urban systems
17
are dynamic and change in the attributes of the objects and
this has its impact upon the system itself and its environment.
Moreover, the growth and change alters the spatial structure
and relationships of the entire system. Though the system
changes alters yet it attains a certain balance which has
given rise to the concept of equilibrium. Equilibrium in an
urban system relates to a state of balance between the system
and its environment, and between the major elements within
both the system and the environment (Berry, 1970).
Urban centers are mainly the places of human dynamic
activities and culture. This new field is the study of urban
ecology (Sharma, 1982) and concepts and generalisation may
be formed regarding their distribution, size, function and
growth. The concepts and generalisations may be related to
the character and intensity of land-use within the urban area
and to the spatial interaction between the constituent parts
of the system (Garner, 1970).
Urban dynamics also "the demographic, economic, and
geographical change involved in the shift of population from
rural to urban is in fact a process of system growth and
structural transformation" (Bourne, 1975).
There are different approaches to Urban Dynamics, as
is exemplified by the works of Paelinck (1974), Moody (1975),
Bannister (1977), Boventer (1978) Beumer and et. al. (1978),
Berry (1961), Morill (1984). Tkachanko (1984), Birkin and
Clarke (1985), Yixing and Qin (1984), Dendrinos and Sonis
(1986), Pumain, Saint and Senders (1984), Thakur (1980) and
Ferrester (1969).
Birkin and Clarke (1985) tried to assess the outlines
of main developments in comprehensive dynamics modelling that
have occured over the last 20 years. They presented a framework
18
that allows for the integration of macro and micro approaches
to 'Urban Dynamics' and examined models of spatial structure
that take account of recent developments in relation to rents
and prices explicitly. Their works also contain the speci
fication of a micro-simulation model of an urban system focu
ssing on household activity dynamics and presents some first
results from an attempt to use the framework related to retai
ning system. Finaly they outlined how the models can be exten
ded into a comprehensive framework and explains how list proce
ssing can be used as an efficient accounting scheme.
Regarding urban dynamics and urban system - Berry (1961)
expressed his view that A city can be viewed as a system within
the larger system of cities". Morrill (1984) found that the
current structure of the American settlement system arose
out of the industrial revolution. Industrialization was the
catalitic agent for massive inter-regional population shifts.
The mechanization of agriculture created a surplus population
in the rural districts of the nation which led to the movement
of rural people to urban locations in pursuit of jobs. Here,
the same industrialization fostered the concentration of econo
mic activities in urban area as a result of corporate savings
accured through economics of scale. The net result has been
a settlement system with extreme urban agglomeration within
specific regions.
A similar view was proposed by Tkachenko (1984). He
studied the physical geographical aspects of urban areas and
also included the impact of urbanization on the environment.
Such an approach would involve a delimitation of the system
and a three-stage investigation, encompassing the basic physio-
geographic conditions in the urban study area, the man-included
flow of matter and energy through the physical geographic
system, and the specific spatial organization of the system.
19
Paclinck (1974) illustrated the urban dynamics as an
alternative methods for the urban study. His article reviews
experience of the study of urban dynamics. The main thesis
is that a combination of all methods is, and will be necessary
to fully understand the complexities of urban life. He has
adopted three mathematical methods for expressing the use of
theoretical models, the simulation of more complex systems
and urban econometric methods. Moreover, each of these models
can, and must, contribute in its own right to the study of
urban phenomena over time. The tested models were, linear static
system, an alternative model 1 & 3, linear dynamic system,
and non-linear static system.
Forrester (1969), Moody (1975) and Thakur (1980) have
made the same opinion on 'Urban Dynamics' model of an urban
area. However, Forrester began to apply the modelling concepts
of engineering control systems to the problems of industry
in 1956. Since then, he has come to believe that these concepts
can be applied to a wide range of social and economic problems.
His consequent publication 'Urban Dynamics' (Forrister, 1970)
is on application to the problems of urban areas.
His 'Urban Dynamics' model consists of a set of inter
related mathematical equations describing the population size,
the number of business, the number of dwelling units, and the
information flows for making decisions. The stock of businesses
in the urban areas is divided into three categories. The model
is used to simulate both the growth of an urban area and the
effects on a mature city of such public programs as workers
training, job programmes, and construction projects. Forrester
opens the cave and it is in the sprit of examinations and not policy
action that 'Urban Dynamics' must be treated. However, all
these comments on the model contained in 'Urban Dynamics' should
not be taken as a rejection of the basic philosophy or techniques
used. So, a new concept in modeling is presented for those
20
who are interested in economic and social behaviour in urban'
areas. Ideally, as a guide for public policy towards city
problems, this 'Urban Dynamics' model would be treated as
a powerful technique of urban analysis.
Boventer (1978) has stressed cycles of 'Urban Dynamics',
for the understanding the 'Urban Systems'. It might be seen
shaped by a combination of changes the micro-environment on
the basis of decisions taken through the inter neighbourhoods,
i.e. on the basis of changes in the macro-environments. It
can be said that such developments are then a combination
of intra-locational (intra-neighbourhood) and inter-location
(inter-neighbourhood) effects. In both cases, he tried to
correlate the dynamic processes with special kinds of external
effects called 'bandwagon effects'. He also tried to analyse
them briefly as forms of 'Product Cycle'.Furthermore, he
tested an integrated view of 'bandwagon effects' and 'Product
Cycles', agglomeration economics and the gravity models and
diffusion models appear as special cases. Same conclusions
have been arrived at by Hartshorn (1980). He also stressed
the environments and distances, agglomeration factors, and
prices, as well as 'expectations about environments and the
role prices play. The environment itself is made up of structural
featural of the urban system as related to different kinds
of 'Stock variables' - people, houses, machines, infrastructure,
air quality etc.
Hartsharn tried to measure some urban dynamics phenomena
through different mathematical models like as (a) land use
changes in the absence of agglomeration economics and Bandwagon
Effects; (b) Bandwagon Effects on quality range, household
buy, and commodity mechanism; (c) urban environments and urban
processes through input and output structures; (d) externalities,
Bandwagon effects in cities: as an important determinants
21
of demand are -prices and costs, incomes, and changes in techno
logies and in tastes; and (e) development and decay, and cycles.
Hartshorn (1980) has expressed his view on dynamics
of land use in urban area. It has that land use dynamics exhibit
four processes of changes:
1. Suburbanization(growth on the periphery);
2. renewal (mainly downtown);
3. expansion of public uses and utilities (parks, express
ways, telephone service, etc. ) and
4. growth or decline of nucleations containing specialized
functions.
Beumer and et al, (1978) have taken 118 basic equations
of Forrester's urban dynamics model and analysed them by forming
them into qualitative structural matrices. Subsequently, it
was found that the endogenous dynamics of the system are deter
mined by only 42 of the 118 stock variables, thus enabling
the majority of the equations to be discarded. In the urban
dynamics model, city growth behaviour is found to depend essen
tially upon an attractiveness- mobility base whichs conditions
labour and unemployment changes rates and these in turn influence
change rates in housing and management and finally new and
mature business growth rates.
The 'Urban Dynamics' model describes the growth process
of urban areas. An urban area is defined by Forrester (1969)
as a system of essentially related sectors of industry, housing
and population, an extremely simplified image of urban reality.
Furthermore the dynamic model simulates urban development
during 250 years distinguishing the following sectors:
1. Underemployed sector
2. Labour sector
3. Managerial professional sector
4. Premium housing sector
22
5. Worker housing sector
6. Underemployed - house sector
7. New enterprise sector
9. Declining industry sector
10. Tax sector
11. Job sector
Here, the investigation started from the model as repro
duced in 'Urban Dynamics'. Essentially, the model consists
of 150 equations. The variables in the model could be distin
guished into endogenous and exogenous variables.
As a general system theory. Berry (1961) has discussed
the urban dynamics factors in his "cities as systems within
systems of cities". He tried to examine that urban settlement
patterns may be regarded as a particular kind of system. All
urban settlements on the earth's surface form part of larger
integrated system. Even the smallest village is part of a
larger system. A city can be viewed as a system within the
larger system of cities.
Pumain and Sanders (1984) have expressed their view
on "modelling of intra-urban spatial dynamics". The model of
intra-urban spatial dynamics has recently witnessed some impor
tant innovative impulses from catastrophe theory as well as
from theories on dissipative structures. A series of theoretical
experiments, coopted with simulations based on observations
from urban history is now, aiming to unravel connections between
the mathematical form of urban interactions and the qualitative
form of urban configurations.
Using the 'Urban System Concepts! a number of articles
and books have been published in different time and places.
Some pioneer works are those of Berry (1964, 19966, 1967, 1968,
1971a, 1971b, 1971c, 1972a, 1972b, 1973), Berry and Horton
(1970, 1974), Berry and et al. (1974), Bourne (1974, 1975,
1983), Charnes and et al. (1972), Chen (1972), Chisholm (1972),
23
Chisholm & Manners (1971), Curry and Mackinnon (1974), Domanski
(1973), Friedmann (1966, 1972) Hoch (1972), Hughes (1972),
Morrison (1972), Pahl (1971), Parr (1970), Pederson (1970),
Pred (1966), and Von-Boventor (1973). And some recent work
have been completed by Barras (1987), Mirucki (1986), Alam
(1984) Whitelaw (1984), Yamaguchi (1984), Dziewonski (1984),
Caves (1984), Jain (1983), Ramachandran (1982), Misra and Bhagat
(1980), Kennett (1980), Simmons (1984), Borchert (1986), Whyte
(1985), Pumain (1984) and Juppentatz (1984).
In western Europe, many early writers have analyzed
the urban dynamics factors and urban system concept. Gras (1922),
Christaller (1933), Losch (1937), Harris and Ullmen (1945),
and Vinning discussed the nature and origin of systematic varia
tions in the characteristics of urban places. During the 1940s
and 1950s Rudyard Vining has written over a dozen papers on
such topics as regional systems and urban growth. However,
the full range of implications deriving from the urban-system
concept was not outlined until much later, notably in the work
of Otis D. Duncan and his colleagues in the publication Metropolis
and Region (1960). It is an extensive statistical compilation
and it puts a new view of the nation's economic and social
landscape.
Perloff et. al. (1960) have supported the views of the
authors of Metropolis and Region. They postulated that both
the national economy and the national geography of the United
States could be described in terms of the urban system. Through
out their book emphasis was placed on the variety of specialized
roles of cities within the national system and on the inter-
dependencies among regions and metropolitan areas.
Bourne (1975) presented a remarkable study on 'Urban
Systems'. At first he defined the Urban Systems with respect
to conceptualization of urbanization. However, his conceptuali-
24
zation of urbanization proposed as the basis of this review
is that of the 'urban system'. Also states this view as an
urbanization— the demographic, economic and geographical changes
involved in the shift of population from rural to urban. Further
more, the cities and urban regions of a modern industrial economy-
constitute a set of interrelated subsystems nesting in a complex
hierarchy of increasing scale upward from individual urban
areas to a national urban system. In the light of urban system
definition, Thompson (1972), Richardson (1973a and b) have
discussed the theoretical and analytical methods and knowledge
of system development for integrated approach towards urban
system changes through inter and intra migration. A detailed
discussion of all the urban systems literature is beyond the
scope of the present literature review section. A brief summary
of some of the key arguments on the definition and operation
of these systems has been attempted in a specific order. This
order has been setup by Berry (1964, 1972 and 1973) and summarized
as consisting of at least three levels:
1). a national system dominated by metropolitan centres
and characterized by a step-like size hierarchy,
with the number of centres in each level increasing
with decreasing population size in a regular fashion;
2). nested within the national system are regional sub
systems of cities displaying a similar but less
clearly differentiated hierarchical arrangement,
usually organized about a single metropolitan centre,
and in which city sizes of smaller order drop of
more quickly than in (1) above as one moves down
the hierarchy;
3). contained within these subsystems are local or daily
urban systems representing the life space of urban
25
residents and which develop as the influence of
each centre reaches out, absorbs, and reorganizes
the adjacent territory. In small country levels
(2) and (3) may be difficult to differentiate,
whereas in larger countries both of these levels
may show further subdivision.
Bourne also stated that the problem of defining the
levels of such systems in reality still remains unresolved.
While the national urban system may be easily recognized the
differentiation of levels within that system is not easy.
Nor are these levels likely to remain fixed over time. Rapid
growth and the spread of urban life styles have blurred tradi
tional boundaries between urban and rural, and even between
small and large cities. Berry (1968), Hall et. al. (1973),
Simmons (1974) have pointed out the traditional attempts at
boundary definitions, for individual urban regions and urban
hierarchies, also brought increasingly into questions. Clearly
no one set of definitions will suffice for all purposes, or
for all time.
In relation to properties of urban systems as social
and spatial systems. Bourne has classified these systems in
three basic dimensions: Structural, spatial and temporal.
Clearly structural refers to the hierarchial or vertical orga
nization of the national and regional economics which constitute
a modern nation.
Dickinson (1956) has tested the centripetal forces,
that could determine the structure and spatial distribution
of towns, but on the other hand the centripetal forces still
determine the character of both "town" and "city", but centri
fugal forces have changed the structure of the urban community.
He also explored the centripetal and centrifugal forces that
area complementary to each other, and are fundamental to the
26
development, location, functions and physical structure of
the urban community in all ages. Jefferson (1931) has postu
lated city systems do not grow-up of themselves. The country
sides set them up to do tasks tha.t must be performed in central
places.
The process of urban growth and urban system changes
can be validated in the general process of urban system growth
and changes that underlies the location, growth, and structure
of cities has been summed up as follows. This process has
been suggested by Robert E. Dickinson in 1956.
1. concentrations; This is the tendency for people
to cluster in cities as near as possible to each
other, to their work, and to the amenities of city
life. It is essentially centripetal in character.
2. centralization; Centralization denotes the "distri
butive pattern of population and institutions in
the area of (urban) concentration, and the process
whereby the patterns appears" (Gist and Halbert,
p. 148).
3. Deconcentration; This term refers to the tendency
for people and institutions to shift out from the
existing urban area to the open land on its out
skirts. It is the results of centrifugal, as opposit
to centripetal forces.
4. Decentralization: Decentralization is to be distingu
ished from deconcentration. The latter implies simply
the expansion of the brick and mortar of the urban
complex. Decentralization, on the other hand, implies
the shedding of certain of the city's activities-
such as industry or commerce and administration
to a distinct and separate town that itself functions
as an independent local and regional center.
27
5. Residential segregation; This refers primarily to
the concentration of residents into districts, similar
to the district concentration of distinct economic
uses. It implies, according to the sociologist, "the
clustering in space of persons or institutions" (Gist
and Halbert, 1948,p. 175).
6. Invasion and succession; Invasion and succession are
terms used to indicate this process of changes in build
ings, their use and their occupants. The process is
one of the displacement of one dominant type of land
use or population group by another. Juppenlatz (1984,
pp. 89-94) has plotted the possibility of a basic urban
information system for any municipality in designed
policy, diagnostic, analytic and optimizing models
can draw the simulated data in quantitative and spatial
terms.
Pumain and Julien (1984) have evaluated French urban systems.
They have investigated that urban networks leads to questions
about modes of regulation and of adoptive behaviour in the
system. The results of disaggregated studies of growth and
socio-economic characteristics of 100 to 410 largest towns
in France emphasises the interdependence of various elements
in the system as a whole.
Whyte (1985) has developed an ecological approaches to
urban systems. He evaluated the achievements and shortcoming
of ecological approaches to urban systems and defined the various
ways in which the widely used term 'integrated ecological approach'
has been used.
Borchert (1986) has stated that urban system is transition
and government action approach. He also added some general
thoughts about the incorporation of government influences.
28
Bala (1982) has been trying to explore urban systems and
newly developing towns in India. Her paper has investigated
the mode of emergence, population size, functional pattern
of new towns in India. She elaborated the following points-
i. an overwhelming majority of new towns were not new
settlements but expanded villages;
ii. mode of emergence was the most vital factor in deter
mining demographic attributes;
iii. new towns represented two contrasting tendencies in
Indian Urbanization:
Concentrated urbanization around metropolitan nodes and
dispersed urbanization in recently developing backward regions.
Misra (1984) has explained the 'urban system' of a 'Develo
ping Economy' in micro level study. He tested and analyzed
the concept of city-region, techniques of delimitation and
their application to the city-region. He also tried to evaluate
the size and spacing of the twelve urban settlements- that "the
law of primate city is more valid than the rank-size rule",
and suggested to narrow down the gap between Allahabad and
other towns in rank-size. He tested the "Planning and Develop
ment" through growth pole approach. For analyzing the urban
system, he also used the christaller's city-region and settle
ment hierarchy.
In the case of urban system and urban Traffic Flow, Ogunsanya
(1985) has formulated a model that could be used in urban traffic
system. He also generalized the urban traffic flow in developing
countries.
Chapman and Wanmali (1981) have discussed the magnitude
of the urban system change in India through Urban-Rural relation
ships. They also pointed out the urban-rural dichotomy.
29
Kennett (1980) has using an urban system framework, found
that the erosion of high female-male ratios in most urban areas
was taking place.
Dynamics of urban growth model has been discussed by
Simmons (1984) with the Canadian example. The growth of the
Canadian Urban system in the past has been highly erratic and
largely unpredictable. The fluctuations in time and space of
resource production for world markets have been described as
a series of regional staple economics where in the rise and
decline of an urban subsystem is determined by growth or decline
of a single primary commodity.
Berry and Horton (1970) have tried to explore the nature
of urban system changes through urbanization and urban environ
ment. They have taken different example from American
urban system. In fact, they start with 'mercantile Beginnings
through Industrial Transformation' and tried to investigate
relationships between the Heartland and the Hinterlands. The
heartland had initial advantage of both execellent agricultural
resources and a strategic location in respect to minerals.
Later it grew into the urbanized center of the national market.
Subsequent metropolitan growth has been in a pattern organized
around this national core region. Consequently, the growth
of the amenities and service sector has been developed. A number
of "footloose" industries developed rapidly. Quaternary sector
also developed and the non-job oriented population started
the interregional migration together with rise in real incomes,
have all served to produce yet another transformation of the
economy and the urban system.
Bourne and Simmons (1978) have stated that the subject
matter of urban systems is extremely complex. They expressed
the numerous disciplines on system of cities, urban growth,
and national urban policy. The process of urbanization, mechanism
30
of development, integrated factors generating economic growth
and social changes have been interlinked with urban systems.
However, during the last few years the meaning of the
term "urban system" is a set of interdependent cities compri
sing a region or nation. It can be examined that the cities
within one country, and how they interest with one another
as a system to create patterns of growth and development,
they are emphasizing the processes of growth and change.
For change of urban system, the urban population play's
a vital role in the city. The nature and composition of urban
population are reflected not only in their demographic and
socio-economic characteristics but also in the range of econo
mic activities in which the populations are engaged. Conse
quently, it is not surprising that there are many approaches
to the study of urban populations from purely sociological
point of view (Weber, 1899) and Alonso (1964), Wingo (1961),
Muth (1969). Clark (1951), and Betty et. al. (1962) have
also focused on spatial distribution of urban population
densities and structural change of urban systems and ecology
of urban environment. However, Weber has investigated the
urban population in a purely sociological dimension. He exami
ned the demographic, socio-economic and occupational charac
teristics of urban population.
With the connection of dynamics of economic growth,
Shafer has tried to formulate the urban economic growth struc
ture through supply and demand curve. Concepts of market
and market competition also discussed through elasticity
machanism. With the context of dynamics, he also mentioned
the elasticity of demand and real estate market, buyers and
sellers view.
31
Losch (1954), Isard (1956), Smith (1955), Pred (1967),
Logan (1966), Hoyt 1939, 1961), Andrews (1953, 1954, 1955),
Alexander (1954), Alexanderson (1956), Roterus and Calef
(1955), Lowry (1964), Batty (1970), Dacey (1960), Morrison
(1973), etc. have suggested that urban economic activity
play an important role or urban system change over the period.
Ullraan (1956) has spatial interaction in the urban
system. He said that spatial interaction provides the means
for the satisfaction of certain needs arising from the loca-
tional separation of producers and consumers. Ayeni (1975),
Clark (1965) and Goddard (1970) have also supported the state
ment .
As an Urban dynamics, social amenities and urban services
have been playing very important role for urban mobilization
and utilization of economic factors in urban center. This
statement supported by Mabogunje (1973). Ayeni (1979) has
suggested that the provision of these different amenities
requires fundamentally different considerations.
Bourne and Murdic (1972) have used the formal spatial
organization of an urban system and have shown that it involve
at least two dimensions. These are the geometrical framework
within which distinctive cultural groups live and interact
and characteristics of social areas as perceived by the
residents. This is stated by Ayeni also (1979) in his concepts
and Techniques in Urban Analysis. On the other hand, functional
organization refers to the ways in which land uses and social
areas interact, and emphasises the delimitation of areas
of economic and social domination by some nodes of the urban
system (Symanski et. al., 1973).
Bola Ayeni (1979) has dealt with simulation modelling
and the urban system. Kibel (1972) in the Association of
32
American Geographers, who defined a simulation as 'a cross
between a portrait and a caricature of reality', is, accord
ing manner for purposes of explanation, manipulation and
analysis. He also mentioned, every work of interaction, both
fiction and non-fiction, as a form of simulation in the sense
that it systematically selects certain aspects of life to
illuminate and manipulate and that it collapses time.
A more technical defination of simulation is provided
by Batty (1972) who argued that a simulation is a process
of modelling in which solution to complex situations are
reached without recourse to mathematics. Social systems are
characterised by many variables which act or interact simul-
tineously.
Orcutt et. al. (1961) said that the central features
of simulation modelling are that they signify a scientific
approach to the understanding of the system of interest.
By continuously changing the variables in the investigation,
it is possible not only to identify the more important ones
but also to generate new knowledge about how the system works.
In this way, simulation modelling provides an understanding
framework as well as a way of monitoring and changing the
system of interest. Simulation models in Urban dynamics have
also been studied by Kibel (1972), Paelinck (1970), Batty
(1972), Berry (1964), Brown (1969), Medvedkov (1967), Wilson
(1970) etc. But the deterministic comparative static model,
the probabilistic comparative static model; dynamic, discrete
time deterministic model; dynamic discrete time probabilistic,
dynamic continuous deterministic and dynamic continuous time
probabilistic models have been discussed by Ayeni (1979).
Moreover, Monte Carlo Approach to simulation had been des
cribed by Hagerstrand (1965). The distribution of population
densities in most urban systems follows the exponential law
given by Clark (1951) as-
33
d = d exp (-bx) X O '
where d is the population density at distance, x from
the city center,
d is the city center population density, and
b is the density gradient.
Furthermore, Monte Carlo simulation techniques could
provide useful insights into the mechanisms of an urban system.
Perhaps a major contribution of the technique is its identi
fications of the mechanisms of urban system.
Markov chain models are the simplest of the dynamics
and quasidynamic models currently in use in urban dynamics
analysis. There are many types of Markov chains and they have
many applications in the social sciences. In urban dynamics
study has been used by Scott (1965), Ginsberg (1973), Goodman
(1961), and Spilerman (1972).
Entropy Modelling has been applied in the generation
of models of urban and regional systems by Wilson (1970), A
brief history of urban modelling is presented by modelling
methods and application by him (1985). Six main concepts are
considered: a) Spatial interaction, b) activity - Location,
c) Optimization, d) Comprehensive, e) Static, and f) Dynamic
models.
Pred (1973) has investigated the delimitation and measure
ment of urban area, dynamics of urban trade and service areas
and urban spheres development. Hagerstrong (1949, 1950); Dahl
(1949, 1954); Kant (1951); Godlund (1951, 1954, 1956), Enequist
(1951); Borgsten (1951); Johnson (1952) have presented similar
view.
Christaller and Losch postulated for a system of cities,
but some of the factors have been tested in intra-urban study.
Subsequently, the idea has been applied for conceptual or predi
ctive models of city growth and structure. Moreover central
place ideas have sparked off a good deal of research into the
34
internal structure of cities (Berry, Barnum et al., 1962; Berry,
1963; Simmons, 1966; Ganer, 1967; Bourne, 1975; Thakur, 1977).
Berry's retail model (Berry, 1963; 1965) of city structure
has expressed the relationship between the number of establi
shments in a trade area and the population and income of the
trade area.
Rowley (1973) has examined the relation between population
growth and agricultural development, industrialisation, rapid
urban growth and Big city development. Linton (1958), Davis
(1965), Berry and Horton (1970) have been supported the relation.
Kinsely (1978) has found two major problems in Marseille
Metropolitian region:
i) rapid demographic growth together with high unemploy
ment, and
ii) the lack of an established manufacturing base. These
are used as a negative dynamic factors of urban deve
lopment .
Sundaram (1977) has discussed the two central hypothesis
of the theory:
i) the growth of a city depends upon its specialization
in urban service functions, that is on its ability
to supply goods and services, and
ii) the rate of urban growth depends on the level of
demand for urban services over the service area.
Although most studies in urban geography show a pre
occupation with internal urban structure, many geographers
are turning'their attention to the number, spacing, and optimum
size of cities. Grossly, Von Thunen's scheme, postulated more
than a century ago, of a single city in a large uniform hinter
land of an agricultural economy but is, of course, highly theore
tical. More realistic is the scheme advanced a quarter of a
century ago by Walter Christaller, who conceived of a hierarchy
of urban centers.
35
Since the appearance of Christaller's work, several other
theories relating to the number, size and spacing of urban
settlements have been formulated. Singer investigated the appli
cation of Pareto's law of distribution and found that for a
whole series of countries the formula expresses very well the
classification of towns according to size. Zipf put Singar's
discovery in a simpler from by stating that the towns of a
country are arranged according to size, the nth town having
1/nth the population of the first. Two articles related to
the rank-size and spacing of cities have been written by Berry
and Garrison (1955) and Stewart (1958). Berry and Garrison
in their article "Alternate Explanations of Urban Rank- Size
Relationship", compare the schemes of Zipf, Christaller, Rashe-
vsky, and Simon, and arrived at a certain level of knowledge
regarding city sizes. Stewart in his article "The size and
spacing of cities", reviews the rank-size rule and examines
some determinants of town spacing. He concludes that the rank-
size rule has no logical basis but breaks down in many areas
at both extremes. He also finds that there is nothing natural
about the shape of the town-size pyramid but that relative
size of towns in different function classes and their relative
numbers vary with the stage of economic development, with a
rise best indicated by the standard of living. Duncan (1951)
& Shindman (1955) have tried to investigate the "Optimum size
of cities".
Schnaiberg (1971), Noble and Dutt (1978), Mitra et al.,
(1980), Beattic and Watts (1983) have found positive relation
ships amongs urbanization, modernization and industrilization;
each affects the other. An urban dynamics factor analysis has
been found in Bourne's "Internal structure of the city" in
1982. He used Demographic, Socio-Economic and other related
variables of urban system change. However, urban geographer
have recently become interested in describing a complex spatial
36
structure of a large number of Demographic, socio-economic
and other variables through 'principal component analysis
(Mahmood, 1977). These underlying dimensions are extracted
from the given set of structural variables on the basis of
inter-correlations among them. The work of Mooser and Scott
(1961), Ahmed (1965), Soja (1968), Berry (1970, 1977), Bourne
(1982) etc. can be referred in this connection.
The complexity of the evolutionary processes concerning
uniform, random and clustered locational pattern requires the
use of quantitative methods, particularly the dynamic analytical
techniques in urban study (Thakur, 1977). According to Hudson
and Fowler (1969) the pattern is determined by the relative
distances or spacings of a group of objects with respect to
one another. Plant ecologists Clark and Evans (1966-72)^ Dice
(1960), Moore (1962) and Thompson (1966) have described and
analysed the distributions of phyto-sociological populations.
Dacey (1961) has applied the method to urban and rural settlement
patterns. Four different near neighbour techniques are in use
today (Tahkur, 1980). Moreover, one is based on the number
of reflexive relationships between near neighbours, while the
others use the distribution of various distance measures between
near neighbours.
Sinha (1970) has mentioned that the rank size rule is
a technique to measure the statistical regularity of different
group of population. Mandal (1981) said that the raik-size
rule is a regular relationship which often exists between
absolute size of an urban center (measured by its population,
not area) and its rank (the center with the largest population
in the region ranking 1, the next largest 2, and so on to the
smallest town). This relationship may be expressed by the simple
formula formulated by Marshall (1974):
P = ^1 n n
37
where, P is the population of the nth town in the last
of urban centers and P, is the population of the region's largest
urban center, i.e. the center which ranks 1 in the list.
Composite index (CI) can be used in urban dynamics study.
Shafi (1984), Bartz (1985), Ostic and Maiarc (1988) and Best
& Kahn (1986) measure the normal distribution pattern through
sum of all rank-coefficient variables. According to Singh (1970),
composite index can be used to measure urban utilization pattern.
Mandal also used in (1981).
In urban dynamics study, a number of methods have been
used to calculate the centrality value. Christaller, Godlund
(1967), Jain (1971), Thakur (1973), Davies (1967), Singh (1962),
Vishwanath (1972), Dutt (1969), Singh (1971), Singh (1973).
Saxena and Tyagi (1975) and Hauser and Schnore (1967) have
used the urban systeu model with different variables of urban
dynamics.
First of all a mathematical model of hierarchy was given
by Martin Beckmann (1958). He started with assumptions of a
uniform plain and a network of market areas for a set of lowest-
level centres providing the lowest order goods. He than hypo
thesized that the relationship of population of a center (P )
and the total population (P^) served by that center, comprising
the trade area population (P ) plus the center's population
(P ) was:
P = B (P + P ) c ' c r'
which can be written B
P = c L 1-B Jr
and B/(l-B) is the urban multiplier.
However, his next step, departing from Losch, assumed
a christaller-type urban hierarchy of constant bifurcation
38
ratio K, although, in contrast to Christaller, built from the
lowest level upwards. For this hierarchy he wrote
P^ = P + KP^ (w-1) tw cw t
So, this equation states that the total population served
by a center of level w, where w increases from a lowest level
of w = 1, equals its own population plus the total population
served by the K centers of the next lower level that it dominates
So, the urban related literature shows that the different
urban perspective studies have been done from various angles
of urban phenomena, but specific urban dynamics studies are
few and far between. Moreover, economic and social point of
studies have significantly multiplied during last two decades
but not the demographic point of view. So the urban dynamics
study is essential for the demographic perspective.
CHAPTER III
STUDY AREA - BANGLADESH
Bangladesh is overwhelmingly rural, densely populated
and extremely poor (per capita annual income US S 122). About
85 per cent of the people live in the rural areas. Economic
development efforts have been mainly.centred in the urban areas
and have greatly benefited city dwellers. There is a substantial
differences between rural and urban income levels. Average
income of an urban household is significantly higher then
that of an average rural household.
3.1 HISTORICAL SETTING;
The Bengal delta, still building, is one of the newest
geological formations of earth. So are the people who call
it their home. An Austro-Asian race is believed to have first
inhabited the area. Subsequent developments rendered Bangladesh
an ethnic melting pot. The Dravidians arrived from western
India and were absorbed. Then came the Aryans, in the trail
of the former. Mongols, Arabs . Persians, Turks, and Afghans
followed. Many were absorbed. Others left their imprint. But
all contributed to the development of a distinct cultural heri
tage for the country with Muslim influence predominating in
it. In the 11th century in Bangladesh a peaceful process of
conversion to Islam began. By 14th century the area was predomi
nantly Muslim and Muslim rule continued till the British took
over in the mid-18th century. When the British Raj left the
subcontinent, Muslim majority East Bengal become East Pakistan
and so it remained until the country become independent on
December 16, 1971.
3.1.1 Origin, evolution and growth of urban system;
The rate of urbanization in Bangladesh however is low
but historically it is ancient. In 3rd centuries B.C. there
40
existed a city- Pundravardhan which was discovered in Mahasthan-
garh in Bogra district and Poncha nagari in Dinajpur district
of Northern part of Bangladesh (5th centuries A.D.) Paharpur
in Rajshehi district in 6th centuries and in subsequent period
Mainamoti in Comilla, Kothalipara in Faridpur, Bikrampur, Savar
and Sonargoan in Dhaka district. These cities developed and
then fell. Most of those cities had been developed from religious
point of view, but some due to economic reasons also. Of all
the ancient cities only Chittagong and Dhaka have an old archi
tecture and but it gradually developed as a modern city (map
3.1). However, after 1947, the urbanization rate has increased
slowly, but after 1949 for administrative reasons, Dhaka deve
loped very fast. Of course, other cities also developed with
some acceleration. In last 4 decades, number of cities, popula
tion and rate of urbanization have been accelarating at compa
ratively faster rate.
3.1.2 Urbanization in the Newly Developing World (Third World countries)
The concept of newly developing or modernizing countries
of the World must also be understood (Breese, 1966, P. 6).
Such countries may have a relatively low, or even a high, tate
of urbanization. There are obviously many different stages
or levels of development which vary from country to country
and from aspect to aspect of the system itself. However, newly
developing societies, almost without exception, are increasingly
concentrating their populations in urban places. In 1920, 4.8
per cent of the population of Africa, 5.7 per cent of that
of South Asia, 7.2 per cent of that«of East Asia and 14.4 per
cent of that of Latin America lived in places of 20,000 or
more inhabitants; by 1975 the respective percentages were 18.1,
17.4, 23.7 and 40.5 (Roberts, 1978, p.3, T-1.1). Here, two traits
of this urbanization deserve attention.
41
BANGLADESH ANCIENT CITY MAP
VANG A Ancient ceunlr|F
Courie ol Ihe oncienl Bfahmoouiro
— Proboble csurit ol tht Titto in ihr I3lh centurjr A.O,
ImporlonI Archocologicol tiir
0 to *e to ta 180 «ii.ci
q 10 40 «o 10 too 'ta K I L O H C ' I * *
Map-3.1 Re]uan/'88
42
First, there is considerable variation between countries and,
significantly, between continents of the newly developing world
in both the rate and extent of urbanization.
Second, the increase in urbanization begins consistantly from
the mid-nineteenth century onward, but accelerates, in most
countries, in the period of world War 2nd.
These variations in urbanization of newly developing
countries are to some extent determining the socio-political,
cultural and economic history of the country. So, these variables
are to be examined in each specific case and the differences
in the patterns of twentieth-century industrialization in newly
developing countries will be understood. On the other hand,
an equally important variable is that of political institutions
capable of implementing industrialization in terms of Labour
and low tariff legislation or in terms of powerful class able
to commit itself to industrilization (Roberts, 1982). Conseque
ntly urbanization in newly developing areas is proceeding in
a different cultural and economic milieu than that experienced
by the developed world (Hartshorn, 1982).
However, it is estimated that the total urban population
in Bangladesh would be around 15 per cent in 1981 census, if
the population in Thana Headquarters (newly defined urban areas)
outside the Municipality locations and other growth centres
were taken in account. The size of the urban population in
all the neighbouring developing countries (except Nepal) is
proportionately much larger than that in Bangladesh (23 per
cent, 21.5 per cent, 28.2 per cent, and 22 per cent in India,
Srilanka, Pakistan and Burma respectively). Moreover, the world's
population is about 41 per cent urban, 70 per cent in developed
countries and 31 per cent in newly developing regions.
Table 3.1 showing the urban Rural population in some South
Asian countries.
43
Table 3.1
Percentage distribution of Population of 11 south Asian countries
by Urban-Rural.
Country
Bangladesh
2 Burma India
Indonesia 2 Iran
Malaysia
Nepal 2
Pakistan 2 Phillipmes 2 Srilanka 2 Thailand
Year
1981
1975
1981
1980
1980
1980
1981
1981
1970
1981
1984
Urban
15.2
22.0
23.0
22.3
50.8
34.2
6.4
28.2
38.8
21.5
17.0
Rural
84.8
78.0
77.0
77.0
49.2
65.8
94.6
21.8
61.2
78.5
83.0
Source: 1. BBS (1981) Bangladesh population census National Volume, pp. 36-37.
2. Demographic year Book, 1981 various pages.
Report,
3.2 GEOGRAPHICAL SETTING:
3.2.1 Location, Boundaries and Spatial Relationship:
Bangladesh, situated between latitude 20 degrees and
27 degrees north and longitude 88 degrees and 93 degrees east.
Bangladesh shares common borders with the Indian states
of West Bengal, Meghalaya, Assam, the Union territory of Tripura
on the West (3718 Km long) and Burma at the extreme southeast
(280 Km long). The Bay of Bengal in the South runs for over
717 Km. Tibet (China), Nepal, Sikkim, Bhutan, Nagaland and
Manipur are close neighbours of Bangladesh. Moreover, the
country lies between the eastern margin of the Indian subcon
tinent and the western fringe of sprawling southeast Asia
(Map 3.2).
44
LOCATION OF BANGLADESH
-30»N CHINA
^ ^
^ x/7
• ^ ^ NEPAL
/*
N ^ ;
. / - /
25°N
BURMA
2(PN
MaD-3.2 Re)uan/'88
45
3.2.2 Area and Population;
Bangladesh is a deltaic land with an area of 143,998
Sq. Km. It is one of the the world's most densely populated
areas. It is estimated to be inhabited by roughly 100 million
(1987). Out of 100 million there are 51.5 million males and
48.5 million females and the growth rate is 2.17 per cent.
The sex ratio is 1060 males per 1000 females. The density of
population is 695 persons per square Km for the country as
a whole (map 3.3). A somewhat similar situation is found in
West Bengal, Kerala, east U.P. and northen Bihar in India,
Java, parts of the Yangtze valley and Szechwan in China and
the lower Nile valley in Egypt. In 1981, only 15.2 per cent
of the population of Bangladesh was reported to be living in
urban areas (BBS-District Census, 1981).
A general distribution of population and urban and rural
population is depicted in maps 3.4 and 3.5 respectively.
3.2.3 Physical Setting:
Relief Features: The country has two main physical
divisions -
1) The vast alluvial plain and
2) The marginal hills in the east and south-east
But this simplicity of form has two peculiar features.
The plain is watered by one of the most remarkable networks
of rivers in the world, while in the surface of the plain itself
there are certain well marked topographical features formed
by older alluvial deposits which rise several feet above the
dead flatness of the plain (Rashed,1972).
The most significant features of the flat plain itself
is the number of rivers which have carved out an interesting
drainage pattern with the help of their tributaries and numerous
distributaries (GOB, 1983).
46
BANGLADESH DENSITY OF POPULATION
1988 lEXCUUOING RIVERS AND FORESTS)
Persons per 1.61 sq.km. ( . I Less «nan 300
I B 300-900
900 — 1300
1300—2000
Over 2 0 0 0
1
....
Map-3-3 Rejuan/ '88
47
BANGLADESH
ISTRIBUTION OF POPULATION
Map-3.i+ Re)uan/'88
48
BANGLADESH
URBAN AND RURAL POPULATION (BY DISTRICTS)
3.000.000 PERSONS
000.000 SOO.OOO
N
23°
2^!
s (Si
\ I
25!
1 <^-f</ ^^,
Vr **jj
k •\
t.
)l S '\
\ \ \
\ 2?!
1 i * 1
- V ' ,
Ws \r
Kilometers ay*
Map-3-5 Reiuan/'88
49
3-2-3 The Plain; A Dominant Feature of Topography;
An overwhelming proportion of the surface of the country-
is a vast, flat, even, alluvial plain, intersected by numerous
rivers, their distributaries, Khals and backwaters and dotted
with bils and marshes. Moreover, it has been formed by the
deep infilling of the eastern part of the Indo-Gangetic trough;
the two stages of the process are visible in the present land
scape, represented by the new alluvium and the old alluvium.
The New Alluvium;
The new alluvium, is quite distinct by virtue of its
vastly superior productivity and because it occupies about
three times the area of the old alluvium. Characteristically,
the new alluvium of the plain gives an impression of absolute
flatness over great distances (brocken only where the old
alluvium rises abruptly to not more than 30 metres above the
general level).
However, a remarkable evenness of the plain can be found
from the fact that many places, though about 170 Km away from
the sea, are less than 10 metres above sea level and the slope
is less than 77 mm per Km (Rasheed, 1972). So, if the level
of the sea water were to rise by 10 metres most of the country
surface would be under water, except the isolated 'islands' of
old alluvium and the hills of Sylhet and the Chittagong Hill
Tracts.
In these plains there are numerous rivers which are of
fundamental importance to the life and work of Bangladesh.
The rivers can be conveniently divided into five systems
{map 3.6).
1. The Ganges, or the Padma, and its deltaic streams.
2. The Meghna and the Surma system.
50
Map -3-6 Rejuan/ '88
51
3. The Brahmaputra's affluents and channels.
4. The North Bengal rivers.
5. The rivers of the Chittagong. Hill tracts and the adjoin
ing plains.
Each of these great rivers carry an immense volume of
water and a great load of silt in a wide bed (Ahmad, 1976).
Here, two different categories of river, the dead rivers and
active rivers. South of the Padma, between the boundary with
India in the west and the Meghna estuary in the east, are
three major right bank Ganga distributories. Each in its are
turn once the main outlet channel of the Ganga, but, with
its progressive eastward shift, successively reduced in stature.
This area is the delta plain and the major rivers from west
to east are the Mathabanga, forming the boundary with India,
the Garai-Madhumati and the Arial Khan. The Arial Khan is
still an active river, carrying a large volume of Ganga water
to the sea. Rivers west of the Madhumati and all their branches,
distributaries and interconnections, may be classed as decaying
rivers (Ahmad, 1976). North of the Sundarban (the delta face),
comprising Kushtia, Jessore, northern Faridpur and nothern
Khulna is the moriband delta. East of the Madhumati, the delta
plain, lying lower, is honeycombed with interconnecting active
rivers which flood regularly and inundate the whole area.
The delta rivers, supplemented by numerous streams,
finally cross the sunderban, the face of the delta, to discharge
into the Bay of Bengal (Khan, 1959). The flood plains are
the creation of the great rivers, the Ganga (called the Padma
in Bangladesh), the Jamuna and the Meghna. The Ganga reaches
Bangladesh at the western extremity of Rajshahi district,
and after forming the boundary between India (Bengal) and
Bangladesh for 147 km it enters Bangladesh just where it throws
off the Mathabanga to the south. Flowing southeast, it receives
the Jamuna near Goalundo and, some 147 km downstream, the
Meghna (Rashid, 1977).
52
The Jamuna-Brahmputra system gathers the drainage of
the Assam ranges and of the eastern Himalayas, areas of excee
dingly heavy rainfall. Some 64 km below the confluence of
the Tista, the river was divided, following the great floods
of 1787 (Ahmad, 1977). The Jamuna, flowing due south, becomes
the main channel, while the 'old Brahmaputra' becomes a left
Bank distributory. The River Burhiganga, skirts the southern
edge of the old alluvium and flows past Dhaka city (Ahmad,
1977).
From the Naga-Manipur watershed (and the Khasia-Jaintia
Hills), the Meghna comprise of numerous streams of which the
Surma and the Kusiyara (branches of the Barak), re-unite to
form the main channel (Rashid, 1972). All the rivers from
the watershed, however, having reached the plain, in true
Bangladesh fashion, throw off distributaries and branches
which reconnect and interconnect to form a most intricate
pattern. The whole source area has extremely heavy rainfall
(Rasheed, 1972). Lower down, the river receives Assam and
eastern Himalayan waters via two Jamuna distributories, first
the old Brahmaputra and then the Dhaleswari, before being
joined at Chandpur by the combined Padma and Jamuna. Idealy,
the Meghna throughout is a big river, but especially after
the confluence of the old Brahmaputra it grows rapidly, despite
throwing out left bank branches to meander across the flat
Tippra plains. It achieves great width, depth, and velocity
even before being rainforced by the Padma.
From Chandpur to the sea the Padma-Meghna is estuarine,
though in fact tides penetrate upriver beyond Chandpur as
for as Ashuganj.
A sizable past of the north Bengal plain of new alluvium
is now cut off from the Himalayan supply of water and its
53
drainage is deteriorating. Due to siltation in the great floods
of 1787, the western north Bengal plain achieves something
of the moribund character of the western deltaic plain (Rasheed,
1972).
The Old Aluvium; Another most important constituent of the
Bangladesh plain is the old alluvium, the remnants of an older
phase of deposition, comprising two major tracts, the Barind
tract in north Bengal, surrounded by the north Bengal new allu
vium, and the Modhupur tract in south Mymensingh and north
Dhaka districts, and one minor tract, the Lalmai Hills, Southwest
of Comilla town. The following are main old alluvium (map 3.6):
1. The Barind Tract: This tract, lying between the Jamuna
and The Ganga crosses the Indian border.
2. The Madhupur tract: This tract, measuring some 113 km
(70 miles) from north to south in southern Mymensingh
and northern Dhaka and some 56 km at its widest,
covers about 4,000 square kilometers.
3. The Lalmai Elevation: The so-called Lamai Hills,
running from Mynamati in the north to the neighbour
hood of Lalmai railway station, though nowhere exceed
ing 12.5 metres in elevation, are a prominent features
in the midst of the dead leveling the Tippera plains
southwest of Comilla town (Rashid, 1977).
3.1.3.2 The Hills: Hills occur along only two parts of the
eastern border, in Sylhet in the north and in Chittagong and
Chittagong Hill Tracts districts in the southeast.
The Sylhet Hills: In this part, all the hills are low. Isolated
hills around Sylhet town rise to 30-60 metres.
54
The Chittagong-and Chittagong Hill Tracts Hills: This is only
extensive hill area in Bangladesh, has ten north-south ranges,
which increase in elevation to the most and rise to narrow crests
which fall in elevation at both extremities. Crest-line eleva
tions of 188 metres are exceeded only in the eastern most ranges,
though individual peaks are higher (map 3.7).
3.2.3.3 Physical characteristics: It is on the plain that the
overwhelming both of agricultural production takes place and
agriculture contributes some 60 percent of total production
and employs about 70 per cent of the total area of the country,
with the new alluvium accounting for almost 90 per cent of the
plain or nearly 70 per cent of the total area (Ahmad, 1977).
The plain has been built by the rivers and its current role
in production is intimately bound with the present character
and behaviour of the rivers. A casual glance at even a small
map suggests their immense numbers ranging downwords from the
three rivers, Padma, Jamuna, Meghna, through great to small
tributories, branches and distributaries and forming close mesh
of interconnected channels.
Finally, the rivers provide a continuous, close network
of navigable waterways but at the same time present innumerable
and serious obstacles to road and rail communication, especially
where banks are constantly shifting and where building of embank
ments can produce damaging changes in river behaviour.
3-2.3.4 Climate and Soils; The tropic of cancer passes through
Bangladesh and so tropical monsoon characteristics are pronounced,
Temperature varies between 9.8°C in January to 31.8°C in July
(map 3.8 and 3.9) and the mean annual variation ranges from
13.9°C to 26.7''C. The summer monsoons are preceded by pre-monsoon
rains in May that often arrive in the company of violent nor-
westers, and are frequently succeeded by cyclonic rainstorms
and thunder showers that may go on well into November. Such
55
BANGLADESH
PHYSIOGRAPHIC DIVISIONS 0 ^0 80 1 0 1^0
Kilometers PLEISTOCENE TERRACES.
Map-3.7 Rejuan/'SQ
56
. \ 1 • •
/^ \2k^^^-^
1 I ill
BANGLADESH MEAN JANUARY TEMPERATURES
( in degrees Centigrade)
BENGAL
toe
Map-3-8 Rejuan/ '88
57
— 1 — ti'i
BANGLADESH MEAN JULY TEMPERATURES
( in degrees Centigrade) JC»LE
40 *o 10 100 Macs
>0
L_
Map- 3-9 Re]uan/'88
58
storms and occasional tidal waves cause considerable devastation.
Annual rainfall varies form 1.27 m in the west to 2.54 m in
the south-est (map. 3.10).
The Geological survey of India compiled a generalized soil
study the sub-continent of India in 1935 (Wadia, Krishnan and
Mukerjee, 1935). According to this study, on the basis of their
geological origin. Bangladesh soils could be divided into three
soil divisions:
1. The newer and older alluvium areas, including north
Bengal, the eastern plain and the Noakhali-Chittagong
coast;
2. The deltaic area (bounded by the Ganga-Padma-Meghna)
or deltaic alluviums. It had three sub-divisions; fine,
silty clays to the north, followed by impure, peaty
deposits to the south, and the deltaic swamps forming
the sea face; and,
3. The sand, gravelly and limy soils towards south Sylhet
and the Chittagong Hill Tracts. Their origin was related
to early Tertiary formations.
In addition, the forest areas in this region were said to have
damp, mixed soils.
3.2.4 IDENTIFICATION OF REGIONS;
Bangladesh can be divided into four major regions on the
basis of Geographical and administrative set-up. The regions
are:
1. Central region (Dhaka division)
2. Eastern region (Chittagong division)
3. Southern region (Khulna division), and
4. Northern region (Rajshehi division)
The regions can be shown in map 3.11.
59
Map-3.10 Rejuan/'SS
60
90'*'E BANGLADESH
REGIONAL DIVISION '/y/\ Northern Region
Central Region Eastern Region
lllll Southern Region
ifw
Kilometers
90tE
Map-3.n Rejuan / '88
61
In relation to finding of the study area the following
historical and geographical factors can be recognised. Geogra
phical factors have been playing a vital role for the origin,
growth and development of an urban center in a area. It is evident
that a greater part of the riverine and wet Bengal delta and
the inland low plains was under forest and/or marsh land through
the pre-historic and even in the historical past. Even as late
as in the Mugal period, much of the Gangetic plain was under
forest and human occupations was quite slow to penetrate into
such lands. There is evidence that major pockets of human settle
ments existed in fertile river valleys of agricultural value.
On the other hand, evidence from population history of this
region suggests that very low population settlement developed
due to number of natural hazards, like flood, storm surges,
sudden shift in river courses etc, which undoubtedly led to
the decline of flourishing human settlements specially in the
southern part of Bangladesh. On the other hand, it leads to
migrate the rural people to urban area for flourishing the urban
population and territorial expansion. The peoples are migrating
from southern region to central region (specially Dhaka Statis
tical Metropolitan Area) and the peoples of rural Noakhali moves
to Chittagong Statistical Metropolitan Area (SMA). Moreover river
acts as an important controlling factor for the origin and deve
lopment of urban centers. The world ancient cities have originated
near to the riverine area and the Bangladesh is not exception
of that phenomenon. In present time, all the big cities have
developed near the riverine area, like-Dhaka, Chittagong, Khulna,
Rajshahi, Mymensingh, Barisal, Sylhet etc. in the bank of Buri-
ganga , Karnofuli, Rupsha, Padma, Old Bramaputra, Lower Meghna
and Surma rivers respectably. High altitude and dense forest
and unfertile lands are inversely related to the urban center
growth and development. Due to high altitude and dense forest
of Chittagong hill tracts have very few number of urban centres.
Modhupur tract and sunderban area are also discouraging for
62
urban growth due to dense forest and water loudging. Barind
tract in northern part of Bangladesh is the sparsely developed
of urban centres due to unfertile piedmont region.
Apart from agricultural aspects physical factors play a vital
role in urban growth and development because it brings out better
differences in linkage facilities. Due to the low rate of road
and railway connectivity - Chittagong hill tracts, southern
part of Khulna, Rajshehi division and north-eastern part of
Bangladesh are bearing a less number of urban centers. On the
other hand, Dhaka, Chittagong, Comilla, Mymensingh, Tangail,
Jamalpur districts have high density of urban centers, only
for better road and railway connection. These high connectivity
is only because of its favourable physical environment.
CHAPTER - IV
URBAN DYNAMISM : GEO-DEMOGRAPHIC AND SOCIO-ECONOMIC CHARACTERISTICS
OF URBAN CENTERS OF BANGLADESH
4.1 An overview of Urbanization:
In most of the countries, the urban population is growing
at a faster rate than in rural areas. People are migrating to
urban areas seeking employment because in rural areas the agricul
tural returns are not sufficient to meet their requirements. The
"demonstration effect" of urban population also helps rural folk
to migrate from rural to urban areas. Most of the industries are
located in urban areas, especially in big cities. These industries
attract rural population to cities, thereby causing overcrowding
and imposing a strain on civic facilities. It has been generally
said that rural poverty and urban misery are the two faces of
the same coin. In most of the developing countries there is a
lot of poverty in rural areas which compels the rural population
to migrate to urban areas.
Urban phenomenon is very old and its origins go back to
the mistry past of prehistory. Until the eve of the 18th century
man's principal habitat was rural, there were very few ancient
cities, which were known for their decisive contribution to civili
zation, they were very few in number and the largest of them all
was smallest by the present-day standard. The growth of towns
and cities, depend upon the basic requirements food, shelter and
clothing. The early towns and cities grew up within heavily popu
lated agricultures areas, or at route crossing (Everyman's Encylo-
paedia, vol. 12, pp. 230-231). Now, dominance of urban system
in today's world is largely facilitated by the accelerated trend
and irreversible urbanisation. Although it is an age-old phenomenon
its dominance is relatively recent, which came in the v-ake of
technological change, economic develoment, development of human
societies, scientific discovaries, rapid communication system
64
and growth of commerce and industry. The tumultuous phenomenon
of urbanization has uniquely revolutionized the traditional,
social, economic, political and physical attributes of the
cities.
The outstanding feature of the present-day urbanization
is the spectacular growth of urban population, transformation
of non-urban settlements into urban settlements and emergence
of new cities. It was believed that in 1800 only 3 per cent
world population was living in towns of 5,000 or more, about
27 million population and which was equivalent to the population
of today's two largest cities. Like Tokyo and New York (U.N.
Paper, 1974). Three per cent became 6 per cent by 1850 and
about 14 per cent by 1900. Presently two fifth of the worlds
population is urban. It is estimated that by 2000 half the
worlds population will be urban.
4.1.1 Definition and conceptsi
URBAN; A large number of definition of urban have been
given by urban researchers. The word 'urban' is derived from
latin 'Urbanus' (Stamp and Clark, 1979). The word infact has
latin root 'civitas' (Chamber's Dictionary, 1965).
However, there is no universally accepted definition
of urban. It has been variously described by people belonging
to different disciplines. Historians view it as a legal entity
and as centers of civilization, political scientists as a form
of political organization, economists as an agency of economic
enterprise, and sociologists view it as having a distinguishing
form of social life (Erickson, 1954). The word urban also has
a somewhat nebulous connotation. As for example, Webster's
New Collegiate Dictionary defines urban as "of, relating to
characteristic of, or constituting a city". Definitions of
urban varies and various authors base their definitions on
different criteria. There are, thus demographic, numerical.
65
occupational and functional and sociological definitions and
those based on presence of municipal and public services, on
administrative considerations and on governmental decisions.
In demographic sense, urban is usually considered, as
agglomerations of a given size. The U.S. Bureau of the census
takes it exactly in the same sense. However, size has not been
the sole criterion for demarcation of urban and rural though
it has been the most widely accepted and the single most important
criterion among several used. The minimum population require
ment for urban demarcation has varied 30,000 persons in Japan
to 10,000 in the Spain and Turkey, 5,000 in the India and
Bangladesh, 2,500 in the United Kingdom, United States and Mexico,
2,000 in France and China, 1,000 in Australia, Canada and USSR,
and 200 in Sweden and Denmark (Hartshorn, 1980).
Some writers like Jefferson (Jefferson, vol. XVI) and
Wilcox (Wilcox 1926) used the term urban for places with a
certain density of population. Thus, in several countries a
density of not less than a specified level, say one thousand
inhabitants per sq. km has been laid down. Sometimes demographic
definitions are based on density on birth rate.
There are numerical definations of urban centers with
certain urban characteristics. The census of India, 1961,
denounces the following kind of places as urban areas:
1. All municipalities and notified areas;
2. all cantonments;
3. all localities though not in themselves local bodies,
yet are part of city or town agglomerations; and
4. all places which satisfied all the three conditions
below at the 1951 census:
a) population exceeded 5,000,
b) atleast three-fourth of the population depended
on non-agricultural livelihood, and
66
c) density of population exceeded, 1,000 persons per
square 1.61 Km.
According to Von Richthofen the urban consists of an orga
nized group in which normally the main occupations are concerned
with commerce and industry as opposed to agricultural pursuits
(Garnier, and Chabet).
Aurousseau defines a urban as a places where transport,
manufacture; import and export of raw materials, education, the
administration of natural affairs are carried on; or, quite simply
a place of residence (Garnier and Chabot). According to Bergel
any settlement is an urban where majority of occupants are engaged
in other than agricultural activities (Bergel). According to
Dicknson urban as opposed to rural implies an activity that is
divorced from cultivation of soil and is carried out in close
association with kindred activities at fixed places - these acti
vities are cultural, commercial, industrial, administrative and
residential (Dickinson, 1967). Davis (1961), Smailes (1963) and
Mayer (1970) have supported the same view.
According to the 1981 census of Bangladesh, the definition
of urban area includes place having municipality (Paurashava),
a town committee (Shahar committee) or a cantonment Board. Under
normal criteria, an urban area must have a concentration of
population of at least 5,000 persons in continuous collection
of houses where the community sense is well developed and the
area is provided with public utilities. Such as, roads, supply
of electricity with street light, water supply, sanitary arrange
ment, etc. An area which maintains urban characteristics but
has a population of less than 5,000 and the thana head quarter
may also be treated as urban area.
Moreover, some growth centers like places of trade and
commercial importance, hats/bazars were also treated as urban.
URBANIZATION: The term urbanization is a cyclical process
through which a nation normally passes as it evolves from an
67
agraian to an industrial society and also the movement of people
to urban areas. In other words, the meaning of urbanization has
been changing with time. It is changing from very simple defini
tions to more elaborate and more complex and refined. In the
older days urbanization was simply the process of people moving
from rural to urban and it only involved a demographic concept.
Today it is measured by many economic and social parameters and
by more refined demographic parameters than the simple increase
in population of towns and cities or increase in their numbers.
URBANISM: Urbanism refers to the way of life generally asso
ciated "to identify the phenomenon of city residence" (Queen
and carpenter, 1953). Bergel accepts it as "a condtion or set
of circumstances (Bergel, 1955)y He mentions heterogeneity, high
degree of dependence of urbanites upon others, the segmental
character of urban social relations and the sophistication and
rationality of the urban people to be the distinguishing charac
teristic. These characteristics differ from country to country
and from one urban place to another in the same countr^^,,,^
The urban way of life may also be evidenced in one's poss
essions; television, radio, telephone, electric devices in the
home, the type of kitchen, the many articles with which the home
is decorated, especially, types of books and pictures (Anderson,
1964). He stated that the machine and mechanisms become more
important. The water-supply system, the drainage and sewarage,
other communication net-work, the street lighting and traffic
control network etc. are regulated by a more precise timing
instrument-like a clock.
TOWN AND CITY: Definitions of town and city by various authors
are based on different criteria. There are, thus, demographic,
numerical, occupational and functional, and sociological defini
tions and those based on presence of municipal and public services,
on administrative considerations and on government decisions.
68
A city can be described as a concentration of people with
a distinctive way of life, in terms of employment patterns and
organization. A high degree of specialized and segregated land
uses and a wide variety of social, economic, and political insti
tutions that coordinate the use of the facilities and resources
of the city make cities very complex (Hartshorn, 1980).
In the case of Bangladesh, a town have a population of one
lakh and above have been treated as cities. Panama describes
an urban center with 1,500 population which displays urban charac
teristics an a town (Garnier and Chabot, 1967).
4.1.2 Origin, evolution of urban centres and growth of urban
population;
Urbanization is an extremely new phenomenon in human
history. The earliest urban centers are called "cities" (Davis,
1961). However, when Paleolithic man moved from his cave into
the shelters he constructed of boughs and leaves, he was making
the first step toward urbanization. Then Neolithic man cultivated
plants, domesticated animals, and introduced agricultue. He crea
ted posssessions in the form of crops, animals and tools, and
possessions bred rivalry, which in turn brought the need for
protection. Families callected into trendly groups and formed
villages in which the agrarian population enjoyed the advantages
of mutual protection (Gallion and Eisner, 1963).
Cities first appeared at least 5000 years ago (Hartshorn,
1980). Urbanization and rise of first urban centers become
possible only after a settled life. Childe has shown how Neolithic
Revolution transformed the economy from food gathering to food
supply and made settled life possible. A second revolution called
by him as urban revolution gasticated in the-womb of time from
6000 B.C. to 3000 B.C. When control over natural environment
was achieved, technical advances and advances in knowledge were
69
made and when as a result of these surplus food production became
possible. This lead to transformation of some rural communities
into urban places (Childe, 1936). Thus the urban phenomenon was
started which was to take different forms in different times
and in different places.
Urban revolution took place first in the fertile crescent formed
by the Tigris and Euphrates river valleys of Mesopotamia near
the ancient shore of the Persian Gulf (Hartshorn, 1980). These
cities were primarily religious centers but handicraft functions
such as tool making, pottery, and basket wearing often also provi
ded an economic base. Eridu is generally acknowledged to be the
oldest of these cities.
Most ancient cities were quite small by modern standards,
typically no more than 1,500-25,000 in population. The few that
did grow to large sizes by standards of the day were Uruk, which
may have had a population of 50,000 and Babylon, which possibly
reached 80,000 persons (Hartshorn). Cities developed in the Nile
valley several hundreds of years after those in Mesopotamia,
Most likely a diffusion process was responsible for the spread
to Egypt. Two important cities in the Nile Valley were Thebas
and Memphis, which flourished around 3000 B.C. (Hiorns, 1956).
Around 2300 B.C. cities also developed in the Indus valley
of present-day Pakistan. The Indian Cities of Mohenjo-Daro and
Harappa were very large administrative-regicAJs centers. Another
ancient city HuangHo valley of China.
The tradition of urbanization passing through Greek and
Roman periods after a lull in the Dark Ages emerges in medieval
period and winding its course through Baroque period in Europe
reaches the industrial age.
In Europe, in the medieval period political unification
of territories according to Mumford (Mumford).
70
This large scale urbanization was achieved because of
a number of factors. It was achieved because of a high level
of agricultural productivity and surplus and the centripetal
influence of steam which is a source of power for industry and
transportation. This became a major force in city-ward migration
of people. Electricity and internal combustion of steam engine
(by Watt in 1965) of the automobile dispersed urbanization far
and wide. Sanitation improvements, centralised sewarage and
waste disposal system, pure water, safe food, prevention and
control of disease lowered the death rate. No less important
was the technological revolution in the urbanization of developed
countries.
4.1.3 Urbanization in Bangladesh:
Urbanization in Bangladesh differs from that of Western
world in three main respects:
a) In the historical past a number of urban centers flourished
in Bangladesh as royal towns, some with religious and
commercial importance which later degenerated;
b) Despite the existence of a number of important towns in
the past, present urbanization in Bangladesh has a colonial
background with an indigenous rural culture; and,
c) Recent urbanization took place in an already densely popula
ted area giving rise to a very high concentration of popula
tion, particularly in the larger urban centers.
Ancient History; The earliest references to urban areas in
Bangladesh are mythological rather than historical. However,
in Bangladesh there were royal towns, some with religious and
commercial importance, and subsequently some with a colonial
background. Some prominent ancient settlements and towns may
be cited here as examples.
71
Till the coming of Muslim rulers in the thirteenth century,
the various main divisions of the country were VARENDRA (North
Bengal), VANGA(South Bengal) and SAMATATA (most of eastern Bengal).
Central Bengal Central Region seems to have changed hand between
these three (see map 3.1).
1. PUNDRA; The Indian sub-continent was to a large extent domi
nated by the 'Mauryan Empire' from the fourth to the second
century B.C. This empire had its base in "the Bihar-Bengal region
and most of Varendra formed the province (Bhukti) of Pundra
Vardhana, with its capital at Pundranagara (present Mahasthan).
From various indirect evidence it may be assumed that Pundranagara
was probably the oldest urban center in Bangladesh. It flourished
with Varendra as a beautiful and prosperous town of the Pundrava-
rdhanabhukti between 5th and 6th centuries A.D. and covered
a large area. There was a good road linking Pundra with other
towns in the west, like 'kajangala' (in Maldha).
2. VANGA: Vanga and other areas were within the 'Gupta Empire'
in the 4th and 5th centuries. Early in the 6th century Vanga
became independent and possibly dominated the whole of Bengal
except the north-west. This may account for the general acceptance
of Vanga-desa (Bangladesh).
3. SAMATATA; Samatata, situated on the Bay of Bengal region,
took part in maritime activities. There is considerable evidence
that traders, merchants and missionaries journeyed to the south
east Asian mainland and the Indonesian islands (Bernet, 1937,
Majumder, 1944, Ganguly, 1943 and Coedes, 1968). It is about
6th century A.D. town.
4. MAHASTAN; Represents one of the earliest city sites of
Bangladesh which flourished during the 2nd century B.C. to 12th
century A.D. It lost importance and declined when the sena kings
shifted their capital from Mahasthan to Deopara (in Rajshahi)
and later to Cauda (in Malda). It is commonly believed that
72
a great flood in the river karotoya accelerated the fall of
the Mahasthan city and of many Pundra settlements (Rashid, 1977).
5- PAHARPUR urban center flourished in 6th century A.D. in
Rajshehi (GOB, 1981).
6. MAINAMATI in the west side of comilla town; Belwa in Dinajpur
about 24 Km of Hili; Bangarh in Dinajpur district about 28 Km
sourth of the present town; Bhagavangonj south east of Dinajpur
town and a buddist religious center referred by jHAuen Tsang;
Bakla a port town of 'chandradvipa' (present Barisal) flourished
during 10-llth centuries A.D.
7. Gokula was a small town near Bogra town; and Banagram was
situated in Kotivarsavisaya at Dinajpur in early 14th century
A.D. (GOB, 1981).
8. Similarly, Sonargaon (east of Dhaka), Savar (West of Dhaka),
Bikrampur (in Munshigonj-Dhaka) and Chatigram (now the port city
of Chittagonj ) flourished on trade and commerce in the Pre-
British period. Some of these enjoyed enormous political power
at that time. On the other hand during 1600 and 1700 A.D., when
Dhaka was the capital of Bengal, Bihar and Orissa, it supported
a population about 900,000 (larger population than of Landon
at that time). This statement is supported by Elahi (1985).
However, most of the ancient towns had declined due to
changes in local physical environment mostly as a result of
shifts in river courses due to which they lost transportation
linkages with other areas and natural protection from external
attacks. Moreover, decline of agriculture due to physical reasons
and the commerce due to war, epidemics, natural hazards etc.
led to the decline of many human settlements and towns in the
past.
Medieval History; As revealed by the recorded history of Bengal,
much of the Mughal rule saw the sponteneous reoccupation of
73
abandoned lands by people in Bangladesh. Such a movement res
ulted in an enhanced river traffic thereby, causing development
of a number of river ports. Some of the Mugal rulers, particularly
Akbar, were keen on the extension of cultivated lands and settle
ment development in waste lands. At the beginning of the British
rule, about 40 to 50 per cent of the total area of Bengal was
settled- the rest was mainly under forest cover and marshes,
and urban centers were few and far between (Elahi, 1985). There
fore, despite the massive size of few individual towns in the
past, the country lacked the 'degree of urbanism' in the true
sense of the term. The proportion of town dwellers was very small
because of the country's weak economic base and limited means
of inter-regional surface transportation and communication. The
existing towns were purely administrative with associated features
in relation to commercial facilities and to river transport,
and most acted as the service center of a vast hinterland of
raw materials to the Hoogly-Calcutta industrial area that was
being developed by the British and due to local capitalistic
partronages (Rashid, 1977).
However, the net result was that even at the end of 19th
century, the total number of towns was less than 20 and they
were mostly administrative and most of them retained much of
the rural characteristics (Elahi, 1985). During this time the
proportion of urban population to the total was very small, below
2 per cent (Chowdhury, 1980).
Urbanization in the present century : During the British period
the importance of towns in this region started to change signi
ficantly.
The towns became the centers of colonial economy and admi
nistration rather than a result of socio-economic development
through an international process of evolution of the region
itself. This stands in sharp contrast to the post industrial
urban development in the European countries.
74
In this section, a limited data have been used to observe
the patter, trends, level and rates of urbanization in Bangladesh,
Regional variation and factors of urban growth have also been
explored. Emphasis however is on demographic aspects of urbani
zation. Generally, urban population growth rate is higher than
the rate of natural increase of population. Growth of urban
population is dependent on factors like natural growth rate,
emigration from rural areas, changes in urban boundary and changes
in defination of urban areas. The changes in the definition
of urban areas and the large scale movement of rural population
to urban centres led to rapid increase in the urban population
of the country.
National Pattern of Urbanization; In table 4.1, the percentage
change in urban population, and rates of growth for the period
1901-1981 have been calculated.
Examination of the table 4.1, reveals the following:
a) the level of urbanization in Bangladesh as measured by per
cent of total population living in urban areas increased
from 2.43 per cent in 1901 to 15.54 per cent in 1981, i.e.,
almost a six fold increase;
b) level of urbanization was very low during the British period
(1901-1947), particularly at the beginning of the century
and it remained almost static during the first two decades
(1901-1911) and (1911-1921). The relatively low urbanization
during the British period may be attributed to deliberate
policy of the colonial government to destroy indigenous indu
stries of Bangladesh and to build up modern industrial-
commercial agglomeration around Calcutta.
Urbanization in Bangladesh received impetus after 1947 when
the region become independent after British rule and became
a separate entity.
75
Table 4.1: Intercensal growth rate of urban population,
1901-81
Census
years
1901
1911
1921
1931
1941
1951
1961
1974
1981
1.
1,
1,
2
6
13
Populat,
Number
702,035
807,024
818,480
,073,489
,537,244
,819,773
,640,726
,273,602
,535,963
ion
Percent of total population
2.43
2.55
2.64
3.02
3.66
4.33
5.19
8.78
15.54
3
7
Change
Number
-
104,959
71,456
195,009
436,755
282,529
820,953
,632,877
,262,361
Ave
Percent
-
14.95
8.85
22.20
43.20
18.38
45.11
135.57
115.75
irage annual
(Exponential)
-
1.39
0.85
2.00
3.59
1.69
3.75
6.62
10.97
Source: Census, 1981.
The rate of urbanization increased during the period
1951-1961. During 1951 to 1961 decade there was a 45.11 per
cent increase in urban population compared to the previous
decade's 18.38 per cent.
Four major factors can be identified for the growth of
urban population between 1951 and 1961.
Firstly the important factor was large scale migration
of Muslims from India in 1947 and afterwards.
Secondly development of new centers of trade, commerce,
Industries and administrative in the Bangladesh region
after it attained a new political status in 1947.
Thirdly Rural-Urban migration.
Finally, natural growth of urban population.
76
In 1974, the country had over 6 million people living
in urban areas, an increase of 135 per cent from 1951. This
phenomenal growth of urban population during the 13 years period
1961-1974 may be attributed to:
i. Some industrial development during the 1960s,
ii. Creation of Bangladesh as an independent country,
iii. Abrupt and dynamic violent political change in 1971.
iv. Availability of statutory rationing in the major cities;
V. Deteriorating law and order situation in the rural
area and wage disparities between rural and urban
areas.
In the recent years, urbanization has moved forward at
a much faster rate. During the 1974-81 period there has been
115.76 per cent increase in urban population. In absolute terms
the population growth has been at the exponentials rate of 11
per cent (table 4.1) annually during the period. This abnormal
growth is to a great extent the result of rural-urban migration
of population. As much as 40 per cent of the total urban popula
tion of 1981 has been estimated to have come from rural areas.
About 30 per cent of the total population increased from the
redefinition of urban areas in 1981. Rest 30 per cent increased
by extension of urban area, inclusion of Thana headquarters
and other hats and hazards with electricity. They accounted
for about 4 million urban population.
According to the 1981 census (Table 4.2), only 13 out
of a total 492 towns/cities can be classified as large urban
centers having population over one lakh. According to rank-size
these are Dhaka, Chittagong, Khulna, Rajshahi, Mymensingh,
Comilla, Barisal, Sylhet, Rangpur, Jessore, Saidpur, Pabna,
and Sirajganj. All together constitute about 63 per cent of
total number population of Bangladesh. Of these again, three
largest cities that is Dhaka, Chittagong and Khulna account
for as much as 48.8 per cent of urban population.
77
Distributional Pattern of Urban Centers in Bangladesh:
Table 4.2 gives an idea of how the urban centers flourished
over time. Number of urban centers almost static during the
period 1901 to 1951, when only 15 new centers developed. The
increase was 31 per cent. But the number of urban centers began
to increase significantly from 1961. These were 78,108 and 492
in 1961, 1974 and 1981 respectively. Percentage increase of
urban centers in 1981 was 925 per cent over 1901.
It also appears from the table 4.2 that 83.4 per cent
of the urban centers had population below 25,000 and 60.2 per
cent below 10,000 in 1981. Percentage of small sized urban centers
become high because of inclusion of Thana/upazila headquarters
as urban.
Regional Pattern in Urbanization;
Bangladesh can be divided into four major city regions.
They are Dhaka, Chittagong, Khulna and Rajshahi Divisions and
they can be designated as central, Eastern, Southern and Northern
region respectively (map 3.11)
Appendix 1 shows urban population in each district accord
ing to three last censuces. The districts have been ranked per
urban population in the census year 1961, 1974 and 1981. As
per ranking of urban population, district Dhaka is at the top
followed by Chittagong and Khulna during 1961, and 1974, but
in 1981 the position of Khulna became 4th. The rank position
of Chittagong H.T. was 4th in 1961 and 5th in 1974 which improved
its position in 1981 by securing the 3rd place. Bandarban district
in 1974 census was 4th but declined to 5th position in 1981.
The main reason behind this improvement in urbanization was
upgradation of hilly areas into urban localities and plane land
population rehabilited into urban locality by giving incentives.
Patuakhali, the least urbanized distr,j g"|| =4s§L61 and 1974 improved
its position to 15th in 1981. ^^^>^'"^^'^'^'^^y^
*r Ace No. "^
00
in
(N (TV
o>
o (Ti
0 N •H
C o
• H -P (D
H
ft o cu c (0
^1
o 00 . o o <-< o
CO O
r o
m o vo o
in o
00 O
in o
o o in o
00 O
w 0) N
•H CO
i n
\ o in
in
f N
(N m
rsi m
(N
• oo o -"T O
m CN •
M"
m CN
•^
o o o
o o
> O
78 o
CN . <Ti O • ^ O
^^ m r--i-H .
CN
,— n r-CN .
' T
— i n f N
^ • a\
CN
^ . r H n •H CN
CN CT\ • CN VD . -1 CN
O 00 . VO TT —1 n
o •
i-H i H
n •
m o CN CN
T
• cTi i n
^ in
vo
CN
CN C^
CN
m
o\ a\
CTi (Ti
O +J
O o o
o in
in
CN
• ^ CN r-( CN
en CN • H ( N
r~ CN
o
in o
o
ON ' J -
O +J
o o o
i n CN
in
O rH
(N m
( j i
o on CN n
CN
t - i vO CN ( ^
o o CN • • I '
cn
CN " *
CO
CN T
(Ti CTi
CN
o o o
CN
vo
f ^
C~~ (Ti r H CN
(Ti
CN
in n
<Ti
CTi
CTi
O -p
o o o
in
cc
vo o r H CM
ro as CN CN
o in CN CN
O CN rH r H
vo
CO 00 r-~ -H • H CN '-i
f N —-
<Ti CN m • <H n in
o
as
00
( ^ vo O O . H CN -H rN
CO
n r o o r- l CN -H (N
CO
o o t H CN
o o o
in
U 0)
c D
a
00
a\ <-t
» w ;3
en c ID U
C
o • H •P (D
. H
a & o a,
x: cn Q)
n <o
i H D i
c ro DD
. (0 0 M <
c (D
X) l-l D
C 0
+J ^1
o a (U (X
> ^ r-00
a\ . H
>_ Cfl CD CQ
«fl 0} 0 u 3 0 C/3
cn u 0)
+ j
c 0) u
c (T3 XI M
a 4-1 0
QJ CTl
in - p
c OJ CJ
u 0)
a QJ
x: - p
cn <u - p IT3 O
• H 73 C
• H
cn •H cn (D
XI - p
c (0 M (U
a c
• H
<U M 3 CD
•H fe
79
The factors behind this trend of urbanization for Dhaka
are largely associated with its establishment as the capital
city. The divisional headquarters of Khulna and Chittagong have
seaports and industrial and commercial centers which attract
people from different districts for better employment opportuni
ties and for trade and commerce purposes. The less urbanized
districts are Faridpur, Comilla, Tangail and Bogra.
Distribution of Urban Population;
Appendix 2 shows distribution of urban population of 1961,
1974 and 1981 by district. It is observed that the urban popula
tion of Bangladesh during 1961-1981 increased by 412 per cent.
The rate of increase in the districts over the period ranges
from 258 per cent in Jamalpur to 1283 in Patuakhali. Per cent
increase in urban population of Patuakhali is followed by Noakhali
(1170 per cent), Chittagong H.T. (683 per cent), Tangail (671
per cent), Sylhet (524 per cent) and so on. Urban population
of Chittagong and Khulna divisions increased relatively in larger
proportion than that of Rajshahi and Dhaka division. This is
due mainly to the inclusion of Thana headquarters which were
not previously recognised as urban centers.
Previously, it was mentioned that Bangladesh is generally
divided into four major regions acording to geographical and
administrative set-up. The central. Eastern, Southern and Northern
region accounted for 30%, 26%, 20% and 24% of the total population
of the country in 1981. Urban population distribution in the
regions were more or less similar till 1961, but by 1981, the
distribution pattern has changed both among and within the regions,
The central region is by far the most urbanized. In 1981,
about 20 per cent of its population were living in urban places
as compared to 15 per cent in eastern, 14 per cent in southern
and about 11 per cent in northern region. The central region
accounted for 39.8 per cent of the total urban population and
the eastern region for 25.8 per cent.
80
Northern region is least urbanized. In 1981 census, it
was only 10.38% of urban population, but within the total urban
population it was 16.2 per cent. It accounted for 24 per cent
of total population of the country. Eastern region accounts for
15.4 per cent of the total urban population but its rank as 2nd
within the division. Percentage of urban population amongh the
region was 25.8 in 1981 census but it was 22.4 per cent in 1974
census.
81
4.2 Nature and characteristics of urban growth dynamics:
The nature, characteristics and process of urban growth
dynamics have been recognized as a concomitant of social and
economic development, but the precise inter-relations have
never been thoroughly understood and the demographic implications
have not been fully analysed. It is desirable, therefore, to
follow the process of urbanization in the various countries
and to relate it with (1) economic indices that reflect the
progress of industrial development in terms of increases in
production, trade national income, etc., (2) indices of social
and political change in such fields as education, health, stan
dards of living, political participation and governmental organi
zation, and (3) demographic trends as indicated by rates of
population growth, family characteristics, mobility of the
population, age structure, size, and composition of the
economically active population, and the like.
The urban growth is in a constant state of change. Every
urban settlement has a high degree of correlation with others.
"Because of these flows, routes develop or decline, cities
evolve, land use change and areas undergo social or economic
transformation" (Ambrose, 1969). Nature and characteristics
of a change in one causes a change in others. Such studies
of growth and change can readily be set within a systems frame
work. Thus, the systems approach offers a viable means to examine
the contribution and significance of temporal processes and
latent order instructuring urban space (Badcock, 1970).
Nature, characteristics and factors of urban growth
dynamics are the results of several processes:
1. Natural increase or decrease, reflecting the balance between
births and deaths; i.e., natural increase of urban popula
tion,
2. Net migration, resulting from the balance between the total
movement in and out of the urban areas.
82
3. 'Push' from the countryside and 'pull' of the city,
4. Growth of new urban units and the increase of size of the
secondary urban centers,
5. Influence of trade and commerce and the development of
new administrative centers.
6. Settlement of immigrants from neighbouring countries (Alamgir,
1973).
7. Redefinition of areas which exhibit urban characteristics
but have population of less than 5,000 and inclusion of
thana head quarters as urban and of. some growth centers
like places of trade and commercial importance, hats/bazars.
Although overall level of urbanization was always low
and still remains low, the rate of urban growth was significantly
high, especially during 1951-61, 1961-74 and 1974-81. Number
of small sized towns increased significantly during 1974-81,
but medium sized towns increased in 1961-74 period.
During British rule, most of the urban centers served
the purposes of collection and export-import for the British
empire. Other urban centers were used as administrative or
religious centers. Actually these were parasitic in character
(Hoselitz, 1954-55). Later many of these centers fluorished
as commercial and industrial centers. Again the importance
of administrative centers gained momentum as educational and
cultural centers, and centers of developed marketing system,
infrastructure, better communication, increased opportunity
of employment, electricity, use of modern technology. During
the British rule, Bengal as a whole served as a hinterland
of Calcutta. The imperial power developed Calcutta as the
principal city to serve as the main administrative center,
commercial entrepot, industrial center important sea port
and a highly organised market. For this reason urban population
was mainly concentrated in Calcutta city.
83
Though urban grov?th in Bangladesh has been vary low
and the growth is of recent development it is expected to play
a "generative" role in Bangladesh (Hoselitz). Urbanization is
not only a function of growth of urban centers and urban popula
tion but also a function of the process of modernization and
industrialization. This is generally reflected in the change
of way of life from rural to urban attitudes. New ideas and
values originate in the cities and are diffused to other areas.
Towns are the centers of learning. The intellectuals concentrate
there. Sometimes cities and civilization are viewed as synonymous
(Crossby). A cross section of people of diversified backgrounds
make up a heterogeneous society. This process of creation and
exchange of ideas, diffusion of knowledge, individual and group
contacts, assimilation of societies and associations contribute
to social change. Urban centers are well equipped with organized
marketing system and infra-structure. This condition is favou
rable for the growth of industries. These factors attract people
from the rural areas and generate a redistribution of population
between urban and rural areas. Demographic transition is also
affected by this change. And this mechanism of both demographic
and economic transition is conductive to urban growth.
Urban areas in Bangladesh attract skilled persons to
some extent and since there are opportunities for new skills
and talents, chance of social mobility is higher there. Urban-
rural differentials exists and hence potential income or life
time earning of a person is expected to be higher in urban
areas. In general, these facilities are not supposed to put
pressure on the existing urban resources. But the cities and
urban areas, already overburdened, are faced with new problems
with the influx of migrants.
Mass influx from rural to urban areas has been a signi
ficant factor of spiraling urban growth rate in Bangladesh.
A great majority of this kind of migrants live in slums (CUS,
1978). Young literate as well as illiterate people move to
the cities in the hope of getting jobs, for higher studies
84
and for business activities. But finding nothing gainful they
become vagrants.
That means that urban growth and the expansion of urban
economic activity are not in balance. Heavy influx of migrants
created new problems like unemployment, a high incidence of
crime, the proliferation of shanty towns and substandard housing.
After the birth of Bangladesh there was a mass movement of
people to urban places in the hope of new economic opportunities.
In 1974 the rural economy suffered a serious set back. Natural
calamities like cyclone, famine, flood and river-erosion rendered
many people homeless and economically destitute. Many people
had to migrate to towns for shelter, security and protection.
These led to the growth of slums in big cities, resulting in
congestion, overcrowding, shortage of food and housing,
unemployment, hindrances to education, medical facilities and
transport facilities.
Bigger towns are provided with better infrastructure,
facilities for the growth of manufacturing industry, attractive
location, rapid access to national and international markets,
external economics, ancillary services. The spread of tertiary
sector also attracts the labour force to the cities. Chowdhury
in a recent study estimated the difference of expected and
actual population in urban areas in Bangladesh assuming the
natural growth rate of population to be equal in urban and
rural areas (Chowdhury, 1983). He found that "quarter of a
million people migrated from rural areas. During 1961-74 the
number stood at 2k million and during 1974-78 about 1 million
people have migrated to urban centers".
There is no precise data on changes in the urban boundries
and, therefore, its impact on urban growth cannot be assessed.
But these changes are too little to have any significant bearing
on the recent rapid increase of urban population in the country
85
(Deputy census commissioner). Designation of new areas as urban
accounted for only 8 per cent of the total growth of urban
population during the 1961-74 period (BBS, 1977), but, due
to the redefination of urban areas, i.e., inclusion of some
growth centers like places of trade and commercial importance,
hats/bazars, it accounted 46 per cent of additional urban popu
lation from the original definition of urban centers (estimated
figures from the 1981 urban census). According to the census
of 1981, more than 50 per cent of the total urban population
in Bangladesh were immigrants from rural areas (BBS, 1981 census)
Using the same source, we find that the immigrants constitute
64 per cent, 35 per cent, 44 per cent and 32 per cent of the
total urban population of the central. Eastern, Southern and
Northern regions respectively. A similar finding was obtained
in another study by Chaudhury and Curlin (1975), Who found
that rural male out-migrants tend to move to large industrial
places (map 4.1).
Davis's has been proved wrong, at least in the case
of Bangladesh. Kingsley Davis (1955) has argued that in most
developing countries the contribution of rural-urban migration
to the growth of urban centers is not as important today as
it was for the industrializing countries of the eighteenth and
nineteenth centuries. Mortality in cities then was substantially
higher than in the countryside and fertility was significantly
lower (Khan and Lawis, 1976), while today improved public health
and medical services have sharply lowered urban mortality rates
and fertility has increased (Davis, 1955). One study has found
that slums of Dhaka, Chittagong cities ( CUS, 1978 ) for the
better communication and above mentioned reasons. In the case
of Bangladesh, immigration continued to play a very significant
role in the urban system change.
Table 4.2.1 shows the number of towns/urban agglomerations.
In 1951 it was 63 but the number of towns/urban agglomeration
86
INTERNAL- MIGRATION PATTERNS ^ ^ Permontnt
• » - Seatonol
>o <io c3 (J -as a a c i
;o «o lo n iQo no >ivO«Ctt*»
U*li
\ \ / / 4 ' ' ' ; Chiltogon
e a r OF BENGAL
Map-i f - i Re)uan/'88
<N
H JO (0 EH
rH 00
1 rH IT) 0^ . H
Q D 0 M O
Q) N
• H (0
>1 JO
C 0
• H •(->
(D M (U C 0 H O) Dl
< C (0
X) ^1
D
c 0
EH
^ in Q) TD 10
H
cn c oa CQ
••
r-l CO CTi i-t
0^ r-H
i H vD ON t—(
.—1 t n <T> <-i
c c
C 1 0 (0 3 -H •H jO Q,+J
o\o
C •H
o\o
1
E D
^^ 0 fd 3 CL, H
c H > H 0 (0 -P
M (U
Z J3
C •H
tAo
C • H
(Ao
1
E D
c C 1 0 03 3 -H
X5 a + j >-i 0 (0 3 a . H
c H 5 H 0 (D -M
U (U
2 XI
C
c C 1 0 (0 3 -H
•H XI a - M
rtP
c •H
o\o
1
E 3 Z
c • H
<A0
C •H
dVo
1
E 3
M 0 (C 3 &< H
10
c H S H 0 ID +J
^ <U
JD
C C 1 0 (D 3 -H
X I a + j V 0 10 3 a . H
w c
- 1 3 -H 0 (D J- '
M 0)
Z XI
w c 3 o
o a 3 O
cn
0) N
•H w
87 n >*
• (N i n
fN 0^
• i H
i - l
ON O i
• i H
rH
o
ro
CO
r-•
( N
n vo
r-( N
•
r-i n
i n
• i n
r-•
• ^
ro (N
i n VO
• i n <-i
o • n
CN
• ON
i n
•
i n 00
t
( N i H
n •
o
n •
^ .H rH
ON VD
• f N rH
i n t
i n
( N
« ( N
ON (N iH
• ^
vO
• . H
r-•
T
O
m cn
CN
in
oo o
CN f
(N
CN
VD
CN i H
• r-ro
t H
• i n
•"S" <-H
VD
o • ON
t H
T
• VD
cn CN
CN CN
•• ^ (M
i n
• o
VO
CN
(N
CM CN
•<r
(3N
00
(N
i n
ON
cn CN
cn
o fN
VD
CN 00
VD
CO cn
o cn
o o
00
VO
cn
i n (N
o o
o o
o o
CN 0^
O O
O O
O o
oo o
o o
o o
o o
00
o o
o o
o o
VD
-— + o o o
« o o i H
M
^ tn . CO a (0 O
H Oi
u >-
M M
tn CO
to H U
ON ON ON
«• CJN ON 0 •p
o o o
«• o i n
* • a 0
04
^-
M H M
(0 CO (0
H U
ON i n i n
^ ON • ^
0 4-> O O O
» i n rH
^ • H 0
cu •—'
> M
(0 to 10
H U
ON i n i n
«> • ^
CN
o +J O
o o
> o - H
« • a 0 G.
^
> (0 CO (0
H U
C3N i n i n
"» CTi
0 +J
o o o
• i n
« • a 0 a. •~^
o o o
^ i n
.—. > c M (0 — x:
-p CO C tn 3 tn 0 <U +J rH
rH « H • (0 a E 0 cn cu
(0 -M O
0)
o > c (0 X ^1 D
c (0
(0 c o
• H +J (0 2
C O CJN
-a c (0
ON
VD CJN
in ON
to 3 to c u
PQ CQ
(1) U U 3 O
C/3
88
went up 78, 108, and 492 in 1961, 1974, and 1981 respectively.
An absolute increase of 72.6 lakh (7.26 million) during the
last 1974-1981 period and it was the biggest urban gain for
any decade since 1901. Since 1951, in all respect class- I urban
areas are gained the highest percentage of urban population.
Even in 1981 only 2.7 per cent urban centers shared 52.43
per cent urban population but small towns were 24.0 per cent
achived only 3.18 per cent urban population.
Appendix-3 shows the growth of urban population in last
3 decades. Map 4.2.1, 4.2.2 and 4.2.3 also indicate the district
wise urban population increase in different catagories.
In case of spatial pattern, 1951-1961 period only 9.5
districts the growth rate between was 100-199 per cent; 9.5
per cent districts are entirely rural characteristics, i.e.,
Bandarban and Chittagang hill tracts. On the other hand 9-5
per cent districts were decreased the growth rate (see table
4.2.2).
Table 4.2.2: Bangladesh: classification of Districts by Growth
rate: 1951-1961, 1961-1974 and 1974-1981.
Growth rate (in percentage)
300 and above
200 - 299
100 - 199
50 - 99
25 - 49
Less than 25
Decrease
Entirely rural
Total
1951-
Number of Districts
-
-
2
3
8
4
2
2
2
-1961
Percentage in total Districts
-
-
9.5
14.3
38.1
19.1
9.5
9.5
LOO.O
1961-
Number of Districts
1
2
8
8
1
-
-
1
21
-1974
Percentage in total Districts
4.8
9.5
38.1
38.1
4.8
-
-
-
100.0
1974-1981
Number of Districts
3
4
10
4
-
-
-
-
21
Percentage in total Districts
14.3
19.1
47.6
19.1
-
-
-
-
100.0
89
. ^ . 85'^ ^°* BANGLADESH '^'
GROWTH OF URBAN POPULATION 1951-1961
S
bj^ A-
\ . ^
0 20 kO 60 h •• I • I i
KM.
26 N-
• • ^ • • \ 3
J - - W - - N / 1 ^ . .
^ „^'~^ •s.-jfcy •» rT:;;
fdV:
r » r . T > i ^ : ; ; : ; : : :
v
-2V
-23*
INDIA (West Bengal)
-22* N
300 and above 200 - 299
'i 100 - 199 50 - 99 25 - W less than 25 Decrease Entifely Rural 9p'E
Map-if.2.1 Rejuan/'8B
90
89 E BANGLADESH 92E
GROWTH OF URBAN POPULATION
26 M
300 and above 200 - 299 100 - 199 —
50 - 99 25 - i49 Less than 25 Decrease Entirely Rural
B A. Y
•* National Average 13 +
9,0 E 9.2 E v
Map t+.2.2 Rejuan/ '88
91
^0 BANGLADESH 92'E
GROWTH OF URBAN POPULATION I97tt-198l
0 20 ifO 60
Rejuan/ '88 Map-tt.2.3
92
Table 4.2.2 shows that 4.8 per cent growth rate has been
observed in 1961-1974 period in 21 districts. But very high
frequency has been found in 100-199 per cent and 50-99 per cent
growth rate group and it was 38.1 per cent district.
In same table, maximum growth of urban district has been
examined in 100-199 growth rate group and the percentage was
47.6. Below 50 per cent growth rate was zero. In case of area,
a rapid urban growth has been found in Chittagong hill tracts,
Noakhali and Potuakhali districts. Chittagong hill tracts received
the high rate of growth due to huge number of plain land people
has been in-migrated through government incentive (see
appendix 3).
Factors Affecting Regional Variation of Urban Growth: The regionl
differentials in levels of urbanization and in the rate of urban
growth tend to be related to regional variations in socio-economic
development. Table 4.2.3 presents some indices of demagraphic,
socio-economic variables by region. It shows that, northern
region is the least developed region of Bangladesh. Whereas
46 per cent of the economically active persons in the central
region are engaged in non-agricultural activities, this is true
of only 19 per cent of non-agricultural activities in the Northern
Region. Eastern and southern region occupied an intermediate
position with 31 per cent and 43 per cent of workers employed
in non-agricultural sector respectively. This is also reflected
in the finding of regional variations in industrialization
measured by number and size of industries. The central region
is industrially the most developed, followed by the Eastern,
southern and Northern regions. Of all the industries in the
public sector, 44 per cent are located in the central region
as against only 7 per cent in the Northern region. The corres
ponding figures for the Eastern and southern regions are 34
per cent and 14 per cent respectively. The regional pattern
of industrialization also holds true even when we measure
93
Table 4.2.3: Selected Demographic, Socio-Economic Characteristics by
Region for Bangladesh.
Variables Total Central Eastern Southern Northern
1. Non-agricultural 34,5 45.9 30.5 42.5 19.1 labour force as percentage of civilian labour forced
2. % distribution of 100.0 44.5 34.0 14.0 6.8 major industries (247)* in Public Sector^ 1981
3. % distribution of industries^ by-number of employees, 1981
Number of employees
500 100.0
500-999 100.0
1000-1499 100.0
1500-1999 100.0
2000-2499 100.0
2500 100.0
4. Per capita availa-
blity generated electricity in KWH^ 16.5 28.52 16.80 14.90 2.60
5. Dependency ratio^ 105 103.2 101.5 109.1 111.1 (per 100)
6. % of population 15 1.95 2.40 1.74 1.70 . 1.60 years and above who have completed education beyond high school level
^BBS,(1987),"Report on urban area", Bangladesh population census 1981, p. 26
Annual Report of Public Enterprises, 1981. ^Bangladesh Power Development Board, 1987. Population 0-4 and 65 /population 15-64x100.
®BBS,(1981) Bangladesh population census. National Volume.
*The figure in parenthesis refers to total number of enterprises.
3 5 . 0 0
4 2 . 3 1
5 0 . 0 0
5 0 . 0 0
4 0 . 0 0
5 4 . 5 5
4 8 . 7 5
3 2 . 6 9
2 0 . 5 9
1 7 . 8 6
3 3 . 3 3
3 6 . 3 6
1 3 . 7 5
1 3 . 4 5
1 7 . 6 5
1 7 . 8 5
2 0 . 0 0
9 . 0 9
2 . 5 0
1 1 . 5 4
1 1 . 7 6
1 4 . 2 9
6 . 6 7
-
94
industrialization by size of industries. Fiftyfive per cent
of the large-sized industries (i.e., with employees 2500
and above) are located in the central region. The correspon
ding figures in the Eastern and southern regions are 2 6 per
cent and 9 per cent respectively. There is no industry of
that size in the Northern region.
Electricity is one of the major ingradients for
economic development. In this respect also, the central
region occupies the highest position and the North occupies
the lowest. The Eastern and southern occupy an intermediate
position. In the central region, the per capita availability
of electricity generated is 28.32 KWH and in the Northern
region it is only 2.68 KWH. In the Eastern and Southern the
respective figures are 16.82 (KWH) and 14.94 (KWH). The poor
economic position of the Northern region is also indirectly
reflected in highest dependency ratio. Education is one of
the strongest correlates of socio-economic development. In
this respect also the Northern region scored the lowest and
the central region scored the highest position. For example,
2.40 per cent of the population aged 15 years and above in
the central region received education beyond high school
level. The comparable figures in the Eastern, Southern and
Northern regions were 1.74, 1.70 and 1.60 respectively. It
is indicated that Demographic correlates, socio-economic
development is in close relationship to urban growth.
Nature, characteristics and process of urban growth,
all of those depend upon the urban dynamics variables, i.e.,
the determinants of urban system changes. The change might
be horizontal, vertical or in inter and intra urban internal
structure of the city system. These changes started since
1961. The above discussion shows that demographic, social,
economic, industrial, commercial, political, transportation.
95
agrarian, psychological, cultural and other urban dynamics
variables are working & bringing change in internal structure
of the city and level of urban growth (Gist and Halbert,
1954). Demographic variables, like migration, fertility etc.
have been playing a vital role in urban system change in
Bangladesh.
96
4.3 Urban Interaction between cities
Urban spatial interaction provides reciprocal relations
between two places on the earth's surface. It is based on the
principles of complementarily, intervening opportunities and
transferability (Ullman, 1956). Urban spatial interaction could
be of at least two types, namely: that which involves physical
contacts like the day-to-day movements of people; and that which
does not require such contacts, like telephoning (Ayeni, 1979).
Sometimes the interaction dynamics depend upon relations both
quantitative and qualitative (Singh, 1976).
Interaction between cities, can be understood through
Gravity Model. The gravity model in the social science realm
has a likeness to Newton's law of gravitation and evolved from
the work of Ravenstein and others (1885) in the field of migration
like Zipf (1949) and Stewart (1948).
Newton's law of gravitation in physics deals with attrac
tion of objects to each other. His discovery states that "any
two bodies attract each other with a force proportional to the
product of their masses and inversely proportional to the square
of the distance between two masses (White, 1963). Newton's law
can be expresed as:
m^ m_ F = K • ^ "^
d^
where, F = Force of attraction between two objects
K = Constant
m^ & m^= masses of body 1 and 2 2 d = distance squared (between body 1 and body 2)
In the social science field, mass is typically measured
by population, reflecting the fact that the greater the quantity
of people, the greater the "force" of interaction (Carruthers,
1956). The quantity of interaction becomes:
97
P P 1 2 I = K -
where, I = level of interaction between two centers
K = constant
P^&P_ = Population of center 1 and 2 2 d = Distance squared (between centers)
The gravitational attraction concept in geographical
research has had wide application in the prediction of
movement between cities. In order to predict the volume
of vehicular traffic or number of telephone calls, the
model suggests that one only need know the population of
the centers and thier distance. Frequently the formula
is simply presented as:
P P 1 2
where
G = Gravitational attraction
P. & Py= Population of city 1 and 2 respectively
d = Distance between centers
By incorporating the influence of size and distance
variables simultaneously, the gravity model recognizes
the greater influence of large cities over distance than
of smaller ones. The model also provides a mechanism for
the quantitative assessment of trade areas around cities
(Hartshorn, 1980).
Empirical use of Gravity Model: Using the formula
discussed above to measure the attraction between places,
one can observe the simultaneous impact of size and distance
as shown in Figure 4.3.1. The figure shows a city of 34.4
lakh (Dhaka), another of 14 lakh (Chittgong), next city
Khulna of 7.5 lakhs and another 2.5 lakh as a standard
Matropolitan cities (Rajshahi). But other 9 cities'
98
population varies between 1 lakh to 2 lakh. The intensity of
flow expected between different pairs of cities can be deter
mined. The greatest level of attraction is between city Dhaka
and Khulna, and next between Dhaka and Chittagong. The least
attraction is between Dhaka and Saidpur due to the greatest
distance. Dhaka and Rangpur have the second lowest attraction
for the same cause but the population rank is 9, among the
13 cities. The attraction between the two largest cities Dhaka
and Chittagong are not the highest because they are also the
most distant (212 km), but the level of activity between them
is higher than between Comilla and Barisal, which are closer
but smaller in size of population (see table 4.2 and
map 4.3.1).
In this way we can measure the traffice flow and telephone
calls through gravity model and can assess the future demand
of the respective subject. Moreover, at the national level
the gravity model can show the expected intensity of flows
over distance. In the case of migration, the distance between
origin and destination cities increases, less movement would
be expected.
BANGLADESH
URBAN INTERACTION DYNAMICS 1988
0 20 kO 60 80 I— I I I —I
Kilometers
25 N^
h23°N
INDEX
• Cities (Population above one lakh)
O Dhaka SMA
9.0°E
Map • k.2.\ Re)uan/1988
100
4.4 URBAN POPULATION DISTRIBUTION AND DENSITIES, AND SPATIAL PATTERN OF URBAN CENTERS
An outstanding feature of urbanization in less developed
countries is the dominance of the primate city, which has a popu
lation several times larger than that of the next city. This
feature is beginning to manifest itself in Bangladesh. Metropo
litan Dhaka has about 4 million people, more than twice as many
as in Chittagong and more than four times as many as in Khulna.
Table 4.4.1 presents region wise population distribution
of last 4 decades and the distributional pattern of urban popu
lation size (map 4.4.1 (1951), 4.4.2 (1961), 4.4.3 (1974) and
4.4.4 (1981)). Among the four region the central region has been
earring the highest percentage of urban population. In 1951,
it was earring 35.9 per cent urban population which rose 46.2
per cent 1974, but the proportion slightly, decreased in 1981
(40 per cent). Second position was secured by the eastern region.
There was twentyseven per cent urban population in 1951 but it
gradually decreased to 1974 but slightly increased in 1981 (25.8
per cent). South and Northern regions are least developed region
among the four. The reason behind it is that the central region
is industrially the most developed, followed by the Eastern,
southern and Northern regions. Of all the industries in the public
sector, 44 per cent are located in the central region as against
only 7 per cent in the northern region. So according to push-
pull factor central region attracted more population in urban
area. If we analyze all 4 maps from 1951 to 1981, the central
and eastern region were marked by relatively dense concentration
of towns. In the Northern and southern regions, towns were few
and far between.
101
MaO''t-'»-l
102
9li —
BANGLADESH
DISTRIBUTIONAL PATTERN OF URBAN POPULATION SIZE
1961
/yyy//^ Northern Region
Central Region
21 N
Eastern Region
Southern Region
8|9 E 9|0 M- 9|2
Map-it.:+.2
103
«oci ^ ^ BANGLADESH
K^^- llrDISTRIBUTIONAL PATTERN OF URBAN POPUCATION IOJOO.OOO
9 0 0 0 0 0
'C^yyj\ Northtrn Region
I I Central Region
[ : s ^ ^ Eastern Region
I TT Southern Rcgioq
ifO 80 120
Kilometers
90E(
Map-if A 3 Rejuan/ '88
104
Map' t t . i iA
105
Table 4.4.1: Urban Population Distribution by Region, 1951-1981
Region 1951 1961 1974 1981
Eastern 4,95,110 6,40,000 14,05,000 34,87,000
(27.2) (24.2) (22.4) (25.8)
Central 6,50,779 10,73,000 29,00,000 53,84,000
(35.8) (40.6) (46.2) (39.8)
Southern 2,79,337 4,30,000 10,49,000 24,68,000
(15.3) (15.3) (16.3) (18.2)
Northern 3,94,547 4,98,000 9,20,000 21,98,000
(21.7) (18.8) (14.7) (16.2)
Total 18,19,773 26,41,000 62,74,000 1,35,36,000
(100.0) (100.0) (100.0) (100.0)
Source: BBS, Census report, 1951, 1961, 1974, and 1981
Spatial pattern of urban centers; After a study of the distri
butional pattern of towns, Author proce to examine the spatial
variations in the degree of urbanization as measured by percentage
of urban population in the total in each district (Table 4.4.2
and maps 4.4.5, 4.4.6, 4.4.7 and 4.4.8). In 1951, only 4.34 per
cent of population was urban. As many as 2 districts were without
any urban population. In 12 districts the urban population was
less than national average, but only 7 districts were more than
national average (see table 4.4.2 and map 4.4.5). Chittagong
district had the highest percentage of urban population (11.78
per cent). Lowest in Noakhali district (1.04 per cent). Both
highest and lowest belongs to the Eastern region of Bangladesh.
It can be noted that highest level of per capita was for Chittagong
district. Major manufacturing capacities and infrastructural
facilities were also substantially high in this district. Most
of the Bangladesh was, thus, overwhelmingly rural until the present
decade.
106
9'0'E I BANGLADESH
•••l^-.>^.
PERCENTAGE OF URBAN POPULATION IN THE TOTAL POPULATION ( BY DISTRICT )
-(
77
D i n a j p u r
\
1951,1961,1971+ AND 1981
0 20 t+0 60 80
Kilometers
a J '•. Rangpur , ' ^
P'"V"*'---^"'-./
V ^ - ' M '•:
i i 21 1 ITTf
.• Jamalpur
'• Bogra
• ^ RAJSHAHI
TW i [?l
m NTangail
A i Pabna /'•' '
p vx <
r 21 «;... i / / Mymensingh ' ;.'
v\ .-•'' r 7
.•-, -.25 N-l
). S y l h e t e . y
)-vJ-\/ f '^:
c
I
Kush t i a
'r...-' •r--^
vx vx DHAKA
r ' ' • • . 1
^ N
- 2 3 N
A'
I// /
Map-'t.'t.S to i+.it.8 Reiuan/1988
107
la
3 a o
s u a
3 a o a
c ra
D
IS H
13
•M C H C 10 -P (D a la l< 3 3 01 a Q. M O
o a
I c ^ (0 "H n 3 C 1: a o 5 o -H
a. j j
I
2^ o
c o
3 H 0 . 4 J O
C .
c 4-1 in C ^ 10
O 3 3 M a 0) >H 0 a. o a
I C 10 ID H
i 3 3 C M A O O O H
O* 4J
t 10 C
^4 O 3 -H
o 0 .
o C H
4J 10 -U C J3 10 la u •-* O 3 3 i4 a (1) M-i O a, o a
n 3 = •g a o -' a. -u
-H 10 10 -i 4-1 3 C o a o H O H
a < 4->
o C H
•»-l 10 -M C J3 10
U 3 3
a) «-( o 0 . 0 0 .
I C 10
n 3 C •K a o
O. 4J
>-{ (0 10 H •M 3 C
o a o t-i O H
O. -M
OJ r w
^ 0 m t£ 10 2
01 •p
(1 •H
l-l 4J m •H
a
in
o o o
r
IN
o o o
o o o
o o o
rg
-H r- m
o o o
o o o
o o o
o o o
00
o o o
o o o
00
oo m
o o o
\ 0 PS) i n w
r-l rH 00
o o o
o o o
oo m O (N
o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o
00 iH n (N
o o o
o o o
o o o
CTV
o o
o o
r^ 00 ir CO
r^ CO - 1 O - I
o o o
o o o
o o o
o o o
o o o
o o
o o o
o o o
o o
o
so
o i n
o i n
so i n
r oo
o o
o o o
o o o
o o o
o o
o Ol
0^ vO
^ ^ o
Oi IN
o o
o 00
o o o
o o o
o o o o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o
. - I in i n fN
^^ rH iH i n
00 o i n 1^ r vo m r ^ vo
f-{ f-i r^ i n
l~
m a\
00
o»
m o o ^ n
rg
r-in
o r
m "(T vO
T <-t
i n i n m ^ ^
rt r- m o
•-I ro
IN -H
o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o o o o
o o o
o o o
o o o
o o o
o o o
o o o
o o o
n J N vO .-I
O O
o <T1
i n IN
CO CO o OS O
in —(
o o
o o
•-I o i n f^ OS 00 ^H .-4 ^H ^ ^H rsi
00 r* i n -^
CO
o
^ 00 in
•
o ON a\ *
(N P-
m
o m vo
• ^
^ fN
O in
o
m f-H
in
in r^ «H
o i n
M IN r~ (Js so
so O 00
i-t
m rH
r « n
so o in
t-t
in
so in
o
in CO r-t
.-^ rsj rH
c 10 J3
u 10 T3 c ID m
X
U
01 c o Ol 10
J 10
10
o z
01 10
10
3
a T3 •H 10 Ct.
J 3 tJl c
•H U] C 0) E >s
10 en c 10
10
•H M 10
m
OJ
o CO (0 OJ
n
10 10
C -M 10 H J3 3 3 (fl 4-1
i : : 3 10 ^£ :<: Q.
10
01
o
3 x ; u a 10 3
• n 10 x : a 10 C M !Ji C J 3 T-i c
•H IT3 10 C Q O , CC X
to u D
< U 2 «t m
108
Areas of absence of urban population comprised of a number
districts in Chittagong region. These were Banderban hill tracts,
Rangamati hill tracts and Khagrachari hill tracts. Most of areas
were hilly and mountainous, inhabited by tribal population,
like Chakma.
Areas with very low percentage, less than national average,
of urban Population were widely distributed over the whole
country. These represented all variety-from the densely populated
central and middle part of eastern region to the sparsely popula
ted Chittagong hill tracts and Rajshahi district areas. The
coastal partof Khulna region fall under the same group.
The fifth decade of the present century was marked by
slight acceleration of urban population growth and continued
growing during 1974-1981. During 1951 to 1961 there was a 45.11
per cent increase in urban population compared to the previous
decade's 18.38 per cent. In 1974, the country had over 6 million
people living in urban areas, an increase of 138 per cent from
1961 but in 1981, the urban population became more than 13
million. It is attributed to some industrial development during
the 1960s and the emergence of Bangladesh as an independent
country.
In 1961, the urban population increased from 1.8 million
to 2.6 million, a growth rate of 45 per cent. The national average
was 5.2 per cent. Bandarban district was still largely rural
in character. Only Dhaka and Chittagong districts were above
the national average with 14.8 and 12.5 per cent respectively,
while all other district were below the national average (see
table 4.4.2 and map 4.4.6). The number of towns/urban went up
from 78 to 108. An absolute increase of 3.6 million during 1961-
1974 was the biggest urban gain for any decade since 1901. Among
the pull factors in urban places, employment opportunities gene
rated by further industrialization were of fundamental importance.
109
The developments under the 2nd five year plans provided basic
metals, industrial machinery, electronics, fertilizer, chemicals
and petrochemicals.
The volume of urban-rural migration would have been
greater but for accelerated operation forces to the contrary.
First: The labour supply from within the urban places
was increasing in association with a consistant
rise in the rate of natural increase.
Secondly: The growing tendency towards commuting in the
wake of new transport links between towns and
their surroundings area also increased rural
to urban migration in Bangladesh.
In 1974, the national average of urban growth was 8.78
per cent. Only 5 districts were above the national average,
They were Bandarban (11 per cent), Chittagong hill tracts
(10 per cent), Chittagong (21 per cent), Dhaka (30 per cent)
and Khulna (15 per cent), but all other 16 districts were
beyond the national average (see tables 4.4.2 and map 4.4.7).
In 1981, the national average was approximatly 16
per cent. Same as in 1974, the same five districts were in
above national average. The highest percentage was Dhaka about
39 per cent of total population, followed by Chittagong (32
per cent), Chittagong hill tracts (30 per cent), Khulna (22
per cent) and Bandarban (21 per cent).
The decentralization of administrative machinery created
new employment avenues at state, district and upazila level
headquarters all through the country. The opening of new
educational and medical institutions and initiation of housing
schemes added to the employment potential of many towns. The
110
Table 4.4.3: Density of Urban Population (District Wise
SI. No.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Name of the Districts
Bandarban
Ctg. H.T.
Chittagong
Comilla
Noakhali
Sylhet
Dhaka
Faridpur
Jamalpur
Mymensingh
Tangail
Barisal
Jessore
Khulna
Kushtia
Patuakhali
Bogra
Dinajpur
Pabna
Rajshahi
Rangpur
BANGLADESH
Area of the districts (in sq.km.)
317.98
745.95
920.60
204.69
242.10
270.50
573.80
261.20
115.40
374.50
114.10
307.60
281.50
346.70
283.10
132.80
100.40
201.40
148.00
297.90
410.80
6446.18
1981*
Total urban population
36.963
1,76,683
18,62,000
5,59,483
4,11,656
4,94,433
38,56,896
3,30,387
2,14,491
6,59,212
1,84,781
5,58,363
4,34,949
9,70,200
3,33,384
1,65,524
2,02,919
2,73,784
3,98,822
5,45,070
7,09,797
1,32,79,897
Density of urban population
116
237
2022
2,733
1,700
1,828
6,047
1,265
1,859
1,760
1,620
1,815
1,548
2,798
1,178
1,247
2,021
1,359
2,168
1,830
1, 728
2,060
Source: BBS, Census report, 1981
*Area wise data are not available of 1951,1961 and 1974 urban centers.
I l l
rise in trade and transport functions in urban places also
enlarged the size of employment reservoir.
Densities of urban population; The density of national urban
population was 2060 persons per square kilometers. Very high
density in Dhaka metropolitan city was 6047 per sq.km. followed
by Khulna, Comilla and Pabna were 2798, 2733 and 2168 respe
ctively. Lowest density has been found in Bandarban where
it was only 116 per sq.km. However, in other 17 district the
density beyond national average (see Table 4.4.3 and map 4.4.9),
Industralization and good communication with all district
is the cause of high density of urban population of Dhaka.
112
^' /
N.
* - • . ' ^
.::i\
r 1/
•A,
1
90 E Bangladesh
DENSITIES OF URBAN POPULATION
1981
0 20 kO 60 80 100 1 I . I . I . I . I
Kilometers
^
y=»' >.
NDIA
N
INDEX
Less than 1.000
1.001 to 2,000
2,001 to 3,000
3.001 and above
rib
r — / - • • - ^
2t+N-
- . I N D I A
3
vi; :\
K^i:::::::::l:\:::::-K
V.:::::
BURMA
9,0 E
Re]uan/'88 Map-t+.tt.9
113
4.5 SPACING OF URBAN CENTER (DISTRIBUTIONAL PATTERN OF URBAN
CENTERS) OF BANGLADESH;
A study of the distributional pattern of urban centers
of Bangladesh, as it appears at the outset, would present no
problem to a researcher since any map showing such a distribution
would immediately bring to light well marked regional variations.
Bangladesh tends to display a more or less uniform distribution
of urban centers.
'Pattern' implies some sort of spatial regularity, which
in turn is taken as a sign of the working of regular process.
There are several terms to denote the spatial relationship among
the towns and cities. These include 'pattern', 'network', 'spacing'
and 'system'. The size and spatial pattern of urban centers is
an important aspect of their locational analysis, and the reason
lies in the fact that the urban centers produce a certain degree
of order in their distributional pattern over the space. Amongst
the studies relating to the number, spacing and size of cities,
the pioneering work is that of Walter Christaller, who formulated
the hierarchy of centers. Since the appearance of Christaller' s
work, several other theories relating to the number, size and
spacing of urban centers have been formulated. Amongst them Zipf's
'Rank-size Rule' and Evan's 'Nearest-Neighbour Analysis' became
popular, followed by a number of researchers in different count
ries- to mention few. Berry and Garrison (1958), Stewart (1958),
Browning and Gibbs (1961), Dacey (Lund, 1962) and Singh (1975).
While examining the causes of distributional patterns,
the identification of any pattern in the spatial distribution
of urban centers in necessary. It may be pointed out that loca
tional patterns in some degree reflect and condition the functional
relations among the urban settlements. A number of models have
been put forth to examine existing pattern (Thomas, 1965). For
example, Gibbs used the following formula to calculated the hypo-
114
thetical distance of nearest neighbour;
Hd = 1.11 / A V N
where Hd is the hypothetical distance,
A is the area of the country (in sq. miles or sq kms.), and
N is the number of urban centers.
King calculated the ratios of 'aggregate', random and
uniform spacing by using the formula:
A R =
1/2 P-1/2
to expressing the pattern of urban centers.
Nearest Neighbour Analysis;
Statistical technique known as the nearest-neighbour
analysis was outlined by Clark and Evans (1954) and exemplified
by King (1962) in his study of the pattern of urban settlements
in selected areas of the United States. Nearest neighbour analysis
indicates the degree to which the distribution of individuals
in a population on a given area departs from that of a random
distribution.
For the purpose of this study all urban centers over 5,000
population were considered. This meant, in the present scheme
a total of 60 urban centers in 1951, 69 in 1961, 101 in 1974
and 329 in 1981 spread over 143,998 sq. kms of Bangladesh terri
tory. The areas have been divided in to four administrative divi
sions of Rajshahi (Northern region), Khulna (southern region),
Dhaka (central region) and Chittagong .(Eastern region).
Method and it's application;
In the application, the nearest neighbour analysis method
is very simple and straight forward. For each division named
115
above a series of straight-line measurements were taken between
the urban centers and their respective nearest neighbour. In
several cases the nearest neighbour of a given town lay outside
of the specified areas of Bangladesh, and they were not considered
as the nearest neighbour urban centers.
The distances to these neighbours were measured and included
in the computation. In cases where two towns were located closer
to one another than they were to any other town, the same distance
was measured twice.
If N be the number of cities in a sample and r the distance
from each city to its nearest neighbour, then the mean observed A Fr distance may be represented as r = ^„ . The mean distance
which would be expected if this population (cities) were distri-E —-—
buted at random r , can be shown to have a value equal to .;—,
^j p
where p represents the density of towns per unit area. In other
way it can be measured, if the mean value for a random distribu
tion, D by applying the formula D „ = 0.5 JA/N where N = number of urban centers
ran
A = area of Bangladesh of weight (143,998 Km ).
The ratio of observed mean distance to the expected mean distance
(R = -^ ) can then be used as a measure of the degree to which n i>£
the observed distribution approaches or departs from random
expection.
Theoritical models are given below:
Dispersion pattern:
Clustered
0
Random
1.0
Uniform
2.15 1
R value n
116
Fig. 4.5.1. The nearest neighbour R Scale
Urban centers Nearest Neighbour
A
B
C
D
C
C
D-
C-Reflexive Pair
Figure: 4.5.2: Measuring nearest neighbours
In a random distribution, R = 1 , under conditions of maximum
aggregation, R = 0 , and under conditions of maximum spacing
(or uniform distribution) R = 2.1491 (Clark and Evens, 1954),
Nearest Neighbour Analysis;
The results of the nearest neighbour analysis are shown in
table 4.5.1. The obtained values of R vary in magnitude from
one area to another and with respect to time dimention, and to
the extent that they reveal regional variations in the distribu
tion pattern of urban centers, they are close to results expected
intuitively or as observed on the map (4.5.1, 4.5.2, 4.5.3, and
4.5.4). R value of 1951 (Table 4.5.1a), in the case of the Chitt-
agong division which displays most prominenty a uniform spacing
of settlement, the result of the analysis (R =1.83), though stati
stically highly significant, is difficult to interpret. The fact
that this area does exhibit a uniform spacing of towns can only
117
be attributed to mere chance. Except Chittagong division, all
other 3 divisions are indicating Random distribution. The R ^ n
ratio is nearest to the random scale. Moreover, the overall pattern
of urban centers of Bangladesh is of random distribution. Here
mean observed value is 30.36 and expected mean distance is 24.49
kms and the ratio is 1.24. It is near to the random, rather to
the uniform scale 2.15.
Same type of pattern has been observed in 1961 and 1974
(see table 4.5.1b, 4.5.1c and map 4.5.2 and 4.5.3). But exception
has been found 1981. Here the total number of urban centers have
increased 329 having 5,000 and more people residing in a urban
settlements. Only Dhaka division shows the random pattern of
settlement, but the pattern of other three divisions approaches
to uniform distribution of urban centers. The reason is the decen
tralization of administration upto upazila level. All of those
upazilas with 500 0+ population have been recognized as a urban
centers. It means all upazila have more or less a uniform spacing
from administrative point of view. Due to this the overall result
of R value approaches to uniform, rather than random pattern
of distribution.
Except 1981, last three decades the randomness in this context
indicates . that forces influencing the distribution of cities
are many and varied. Yet, the influence of transportation routes
including river transport of Bangladesh appears to be conspicuous
in all the areas in which the distribution of towns is random
(see maps 4.5.1, 4.5.2, 4.5.3 and 4.5.4).
118
.0
{ ^ ^ .
T 1 • \ BANGLADESH
DISTRIBUTION PATTERN OF URBAN CENTERS
- /
A. 1951
0 20 itO 60 80 100 1 I . I . ( . I . I
Kilometers
2ifN-
• Town over 5,000 Population
9,0° E j ^
Reiuan/ '88 Map-if.5.1
199
A BANGLADESH
('
i
^" ^\'^^^\ DISTRIBUTION PATTERN OF URBAN CENTERS
i \
• r
1951
0 20 itO 60 80 100 !- I • I . I . I . t
Kilometers
s'-^-—
C
.J
s.
INDIA
• Jown over 5,000 Population
9,0°E
MaD-it.5,2 Rejuan/1988
120
BANGLADESH
^ • \ > - , ^ DISTRIBUTION PATTERN OF URBAN CENTERS
< . • . } 0 20 1+0 60 80 100 ( \ . \ . \ . \ . \ .
Kilometers
• Town over 5,000 Population
9,0 E
Map-^.5.3 Re]uan/'88
121
< ; > . > •
I BANGLADESH
•^' V^ i>^ . DISTRIBUTION PATTERN OF URBAN CENTERS A.
0 20 kO 60 80 100 • 1 ,
• Town over 5,000 Population
9,0°E \
Re)uan/'88
Map-it.5.i*
(0 •H
m
(0 c < 3 O
XI
s: CD
•H <U 2
(0 0)
10 0) 2
ID •
^ 0)
H i3 (0 E-i
i H
m O i i H
•• ^ (0
c — 0 C
•H 0 CO - H
• H t j ) > 0)
•H CC Q - ^
OJ U 3
- p
ro 2
•M
C V (1)
-t-J
-t-l
(0 a
u 3 0
in X3 0) M (0 0) 2
c (0 0)
s: -o <u + j
u <u a X
x; D1
•H 0) c
c • H
CO
u •H +J (0
•H -P (0 •M CO
-t-i O
•H M •P CO
• H
(U T3 0 c (0
• I - l
CO • H
u -o
c (0 (U
s
>. +J • H
w c <u Q
M-l 0
(0 (U i-l <
4-1 0
• 0 2
"O (1) > ^1 (1) CO XI 0
c 3 0
-l-J
14-1
0
~ c
E 0
< ^
c
u
e -a j>i c 10 u
Q) u c (0
+J CO
•H
• E
X •
cr CO
u (U a X CO
c H
-< * * i
• E
0) ^ 0 TJ
•H (0 CO H
• H
> • H Q
(0
c s 0
EH
Cn c (0 m
. CP CO
c •H
w
OS
00
VO
in
m
(N
122
<7^ O
(N
a: a
in
in
<yi
00
(N
00
T t-H
• f N ro
(7v 'S'
• " f N
T
• • ••a*
(N
vD fN
• vD (N
O fN
• in fSI
fN 00
• 00 in
^ n
• o m
r-o^ 'S'
o o o
r-• •<r o o o
( N in in o o o
rsi T f N O o O
r-ro fN
'a' n
•* r~ in n ro
i-H
r-r-o n
00 1—1 T i n
•
c o
•H cn <u K
c •H in X <U (0 X
X -M CO M
• n 0 (0 2 OS —
c 0
•H cn <u Di
c M a)
(0 X c
H 3
X Ui
+ j D 0
C/3
—
(0 ^ (0
X Q
—» c 0
•H cn
H (0 V 4-1
c (U
U —
-. c 0
• H cn 0)
cno: c o c cn i-l to <u -M -P 4 -1 CO
•H fO X W o —
K
fN
O O O
00
n
o 1 ^
X CO OJ
(0 H cn c <0
m
CO o:
00
o
VD
ID
m
fN 00
tK a <
o o
00 0^
c 0
•H en <u OS
c •H ^ n o (0 JC
s: A-> in s-i
•r-i 0 (0 2
(D C H D x :
c 0
•rH cn <u OS
c
u Q) x: +-I p 0 CO
(0 ^ (a x :
^ c 0
•H D1
rH to 1^ 4-1 C (U CJ
• — •
c 0
•H cn 0)
c n K c o c cn 1-1 to 0) +J -U -M W •H (0 £ W
a >- ui — Q - - u -^
123
o 00
• iH r\i
a^ in
• TH fN
m iH
• o (N
^ O
• CO CN
• 00
• (N fNl
t-i vO
• n rsi
vo rH
• m CN
i n CO
• r-t (N
i n
^ *
CO •<:r
r~ (N
• <n (N
vD fN in o o o
vD m i n o o o
r~ <-i vO o o o
00 1—1
cn o o o
cn r-'J' o o o
r-ro rsi ^ ro
^ r--in n n
i-H r-r-o n
CO »H "V i n •«y
CO (T\ a\
-
oc
00 (N
0^
w (D 73 (0
H cn c (0
CQ
00
vo
in
CM
0
Oi
(N
o
in
CK a:
o
.—I i n
« 00
o r-
• 00
00 00
• ^
§
CN
00 00
• 00 iH
<Ti n
• r~-rH
(Ti CO
• r-.—1
o o
• a\ CN
O n r~ o o o
i n I-H
r-o o o
r-r~ CO o o o
o in in o o o
r-n CN • ^
n
• ^
r--i n t^ 00
rH r^ r~ o m
00 >-H 'd' i n • ^
i n CN
c o
•H Ol <U a:
c •H M j a o; to s:
x: 4-1 in u
• n 0 (0 2 CC - -
to C H D x : :«;
c o
•H cn 0)
Q:
c u (U x : •i-i p 0 in ^-
10 ^ i to x : Q
—^
c 0
•H cn (U
rH to M
• M
c (U u ^
,—^ c 0
•H cn
cno: c o c cn M to <u +J -IJ -M m •H to x : u u > -
CO < < cc
in (N
VD in o
cr>
in
n r-n OJ ^ ro
'T r~ i n n n
o n
CN en 00
D <
CO CO
(NI
fN as
m
i n
i H
O «
i n
fN T
• o
r-^
• 1—1
fN
r~ o CO (N o o
•<r o CM fN o o
<N ( T i
CO fN O O
vO f-i
CTi r - (
O O
00
. - I
in
CO
00
o •H C75 OJ K
C •H > x : Q) (0 x; x: -M W 1-1 • n O (0 2 tx ^
0 •H Cn (1) OS
c V4
o (0 x ; c i H
;3 £ ;:<::
•l-J
p 0 en -^
(0 ^ (0
£ Q
c o •H cn
rH (0 ^
4-1
c <u u ^'
c 0
•H Ol
Dice;
c 0 C CD U (0 (1)
•l-l -M -M M •H to
x: w u «
x: in 0) n (0
•H <ji c (tJ CQ
124
<
in
in
in 00 fN fN o o
00
u O
M-t • H C
cn c •H x; u o u a a < II
C o
•H • M
XI •H M +J
•H TD
E o -a c (d a II
OS
(1) •M o z
125
4.6 URBAN HIERACHY:
The concept of hierachy shows the relationship among
cities to the entire distribution, usually measured in terms
of size and functions (Hartshorn, 1980). The concept of hierarchy
relates to Christaller's spatialial model of central place
(Christaller, 1933). It is to be said that a hierarchical unit
(city)" has two attributes that should be distinguished: its
relationships to a higher order unit is one part to whole, to
a lower unit of whole to part. That is, going up the hierarchy,
wholes are divided into parts and separated out. Thus, despite
the overall characteristics of control, each hierarchical level
has autonomy over orders below itself, while being a dependent
on those above (Lukermann, 1966).
Urban hierarchy is expressed as the ranking of cities
into successive groups on the basis of the number of functions,
or the size of the population served by the function, or by
the area of the population served. The idea is that there exist
discrete classes of central cities/towns with associated groups
of functions organized together in a nesting pattern. In this
section of this chapter it will be ivestigated through some
urban variables, whether such an hiearchy exists in the cash
of Bangladesh cities/towns or not.
Development of Hierarchical concept;
The concept of urban hierarchy was first developed by
Walter Christaller. He recognized urban centers as central places
providing a wide variety of services to the surrounding areas.
These surrounding areas were named as "complementary regions".
They were identified as hexagonal in shapes. Variations in the.
size was found with the variation in the size of the center
as large-sized centers provide a number of higher order functions
in addition to those of low order which exist at smaller centers.
The smaller centers and their service areas thus "nest" within
126
the service areas of larger centers. According to Christaller
the distribution pattern depends on three principles. If the
system is determined by the maximum accessibility to the market,
the number of centers follows the rule of threes, that is, the
number of centers in each group from the highest to the lowest
order follows a sequence of 1, 2, 6, 18, 54 etc. If the system
is determined by the transport or traffic principle, the patterns
are linear and nesting follows the rule of fours. If the system
is derived from the administrative principle, the hierarchy
exhibits the rule of sevens.
Losch (1954) confirmed many aspects of hierarchy and
stated hexagon as the most advantageous shape for market areas.
Further modification of the theory came from Berry and Garrison
(1958), who reformulated the theory in terms of "threshold"
and "range" of goods. On the other hand or significant contribu
tion towards the concept of urban hierarchy was of Vining (1955).
He regarded the classification of centers into discrete groups
according to their function as merely a convenient mode of expre
ssion. Compared with these works, Backmann (1958) demonstrated
a continuous rank size distribution in Christaller's concept
of urban hierarchy. Thomas (1961) attempted to reformulate
Central Place Theory in such a way as to remove its dependency
on "typical size classes".
For measuring the urban hierarchy both qualitative and
sophisticated quantitative techniques have developed with the
passage of time. Under assessment of goods and services, computa
tion of centrality index is most popular. Recently refined methods
of multivariate analysis have been introduced. Here the hierar
chical classification can be shown through composite index and
inter variables relationship in the correlation matrix.
127
Methodology of hierarchical classification of Bangladesh: For
the decentralization of administrative function, government has
taken a new policy to grade all upazila headquarter with a popula
tion of 5000 or above as urban centers. Due to that at present
there are 492 urban centers in Bangladesh. In order to obtain
an over all view of urban hierarchy 25 variables and were selected
which were subsequently combined it into 15 categories under
the heading of social and demographic indicators. Then a set
of composite index was computed to show the real situation of
urban hierarchy. There is lack of urban centers data in district
censuses and reports on urban area (B.B.S. Vol. 1-21 and urban
report, 1987), therefore. Statistical Metropolitan Area (SMA)
and Municipalitis have been chosen for the urban dynamics study.
The present study is based on secondary and tertiary materials,
instead of primary data.
Selection of social and demographic correlates for ranking
of urban centers depended a good deal on the degree of economic
development in the areas. Moreover, administrative, cultural
and social, as well as the entertainment habits of population,
are influenced by the type of indicators chosen for the urban
system changes. For example, in a developed country, three or
more banks may be a good indicator of an urban center but in
a developing countries, even one bank may serve the same purpose
(Smailes, 1944).
All four SMA and 71 Municipalition have been considered
for the determination of hierarchical system of Bangladesh, because
data of the newly created urban centers are not available.
A set of 25 variables have been covering both social and
demographic side. These variables are comprise 11 broad groups
of functions, namely: Education, technical education, health.
Road, postal and telephone communication, recreation, urban growth,
literacy, non-agricultural population and total population.
128
The Weightage of Variables; For the judgement of the weightage
of the variables the most serious weakness in hierarchy determinant
research is the lack of a common denominator which may be used
to award weightage of different functions according to their
importance.. In fact, the hierarchy is positively affected by
the status of functions of indicators. If the functions are
not given their appropriate weightage, the urban centers classified
with a adequate status is not possible. Some Indian geographers,
like Pandey (1973), Singh (1979), have assigned the functions
arbitrarily in the order of 1, 2, 3, 4, 5,.... In this procedure
a primary school and a bank have been given identical or subjective
values, i.e., one. Obviously, bank is a function of higher order.
Hence the given values are not of adequate or appropriate status
in relation to the services offered. In a slight improved method
Jayaswal(1973) used the Godland's alternative method, the size
and economic activity of the urban centers, was adopted for deter-
ming the discreteness of level.
The centrality method has been applied for grading the
urban centers. The centrality of urban centers was calculated N X 100 with the help of the formula C= 5 . where c is the required
centrality, N is the number of persons employed in manufacturing,
trade and commerce, transport and 'other services' in a urban
center and P is the total number of persons in the region engaged.
Moreover, among foreign geographers. Grove and Huszar argued
that the status of a town is indicated by the presence or absence
of certain 'key' (Binary system i.e. o = absence and 1 = presence
of that facilities) facilities; they applied a 'points' scoring
system, whereby each facility is awarded like the previous one
as 1, 2 and 3 points and so on (Grove and Huszar, 1964). Abiodun
(1967) found the values in the progression of 1,5, and 10,...
This system appears comparatively better as compared to the former.
She did not discuss the base of the system.
129
Sharma has tried to takle the problem in a more refined
way (Sharma, 1981). He worked out a statistical formula for
the same purpose. The formula was based on the principle that
greater the scarcity of functions greater its importance in
terms of centrality and, therefore, higher the weightage. But
this principle does not hold good in all areas and in each
functional class. In the area under study, there are some func
tions which shows less frequency and their actual importance
is also less. If this formula is applied it will give higher
weightage to a function having far less importance. Through
binary system, a 'threshold value' for the district level
planning has been worked out by Bhat, et. al. (1976) and Wanmali
(1987).
However, it is possible to allot weightage among similar
type of functions. But when different types of functions are
performed, the weightage may be of equal importance. For example,
one can say that a degree college is more important than an
intermediate college in an area or vice-versa. But one cannot
say that a bank is more valuable than a hospital. The consumer
may need both the services at the same time. A person may
draw money from the bank and then avail the facility of a doctor.
Thus, if on this basis functional hierarchy is derived carefully
than a hierarchy of settlements may emerge. To give proper
and due weightage to the levels of functions, 'population thres-
shold' was adopted by different researchers. King (1969), Berry
and Garrison (1958) have applied the 'threshold population'
technique for the ranking of urban centers.
The threshold population is the minimum level of support,
as measured by number of persons required to support a function
in a particular place. Hagget and Gunawardena, and Berry and
Garrison have used two type of threshold population formula.
For instance, Hagget's formula is applicable where the function
occurrence ranges from zero to 100 per cent in all the group
of settlements under study.
130
However a most common and traditional method for calcula
ting threshold values is the means of statistical average,
I.e., X = — N
where, X = threshold value
X = total population of the given center
N = total number of the functional outlets of
the certer in question.
For the weightage Bhat used the formula in following N r-
way: Wi = -=. and Ci = }_ VIx Xij , ^^ i-.i
Fi = Number of settlements having the function/sub-
function
N = Total number of settlements
Wi = Weightage to the ith sub-function
K = Total number of sub-function under a given
function
Ci = Composite value for that function for jth sett
lement
Xij = Value of the ith sub-function in jth settlement.
So, both the Bhat and Wanmali method of threshold popu
lation are based on binary system (that is the facilities exists
or not) of the respective settlement.
To avoid the all crisis, on the urban level weightage
system, a unique method has applied to known the urban level
basic facilities are more than one unit and considered not
ubiquitous in all urban centers. Variables (sub-functions)
are not of equal important and so relative importance and weigh-
tages were assigned accordingly. The lowest sub function (all
Bangladesh average) has been assigned unit score of one and
the weights of higher order sub-functions are obtained by divi
ding the lowest sub-function with the higher order sub-function.
131
In formula:
CI = rji X K
"f
Where CI = Composite Index
L^ = First lower order function.
H^ = sequential higher order function
K = constant factor
Say for example, the number of student enrolled per
lakh population (all Bangladesh average) in primary schools
is 11,695, it is assigned unit score of 1. The next higher
order function, i.e., enrolment in high schools, is 2,681 per
lakh population. The weight of high school enrolment is obtained
by dividing 11,695 with 2,681. Thus, the weight of high school
enrolment is 4.4. Similarly, the weights of other higher order
sub-functions were obtained accordingly.
After obtaining the weights, the corresponding figures
were multiplied with their weights and composite index has
been calculated and presented in appendix-4. Further the indices
are calculated by taking the all Bangladesh as 100. The overall
social facilities have been obtained by adding the 'indices'
of 11 social facilities, the column of indices 1 to 11 and
demographic determinants column 12 to 15 and over all ranking
of the 75 urban centers of Bangladesh has been obtained in
appendix-5. Author has made the above formula and where the
irregularities occured, they have been adjusted through arithme
tic average or ratio/rate with the help of preceding and suceed-
ing values, and where there single unit variables have taken,
those were simply transfered into per lakh population or as
a rate and proportion of the respective variable.
Subsequently, with 11 x 11 correlation matrix of social
and demographic indicators have been calculated manually. This
table 4.6.3 will help to examine the impact of demographic
132
correlates on other urban dynamics variables. However, the
weights for the variables of the composite index are given
in table 4.6.1.
Table 4.6.1: Weighted Scores for the Variables
SI. Major Services and Levels Weightage Facilities
Educational 1. Primary 1 Institution 2. Secondary 3.183
3. College 11.902
4. Technical educationl44.95 like, medical and Engineering College, Universities etc.
Educational Teachers
3. Health (Medical)
4. Students
Post Office
1.
2.
3.
1.
2.
3.
1.
2.
3.
4.
1.
2.
3.
Primary
Secondary
College and Technical education
Primary Health centers and Dispensary
Union Health and Family Planning centers (UHFPC)
Upazila and District Hospital
Primary
Secondary
College
Technical education students
Branch Post Office
Sub-Post Office
G.P.O.
1.
1.
1.
1.
1.
3.
1.
1,
5,
94,
1,
1,
3,
,0
,14
,79
,0
,15
,20
,0
.19
.36
.18
.0
.25
.58
133
7.
8,
9.
10.
11,
12.
13.
14,
15.
Telephone
Bank
Cinema Hall
Transfired into per lakh population through the applied formula in the text.
Ditto
Ditto
Crime Ditto
Proportion of pacca road in respective town/city
Literacy rate 5 years and above
Non-agricultural population
Urban Population growth rate
Physicians
Total population
Per thousand Population Per hundred Population
Per thousand Population per year
Per lakh population
Absulute figure
Source: 1. BBS, District statistics. Vol. 1-21, 1987
2. NILG, Municipality Statistics, 1987.
3. BBS, Report on Urban Area, 1987.
The scale of Functional Hierarchyi The above mentioned scale of functional hierarchy is intenpretated in terms of the relative importance of services. A high school is relatively of higher importance than a primary school, and an University is of higher importance than college. Therefore, the higher the level of functional hierarchy, the higher would be the composite values of the place performing that function.
The classification:
The composite indices of the individual 75 urban centers
of Bangladesh (SMA and Municipality) are plotted on a graph
as social, demographic and total composite values:
134
i) Whether there were natural break points on the graph
that could be used as reference points for classifying
the various urban centers into separate classes, and
ii) Wheather there was any correlation between the total
population size of the urban centers and social variables
of composite index and in correlation matrix.
The urban centers range in population size from 8 thousand
(Mahespur) to 35 lakh (Dhaka SMA) and it was plotted in a
simple graph. The population was plotted on the Abscissa (x-
axis) and social, demographic and total composite index has
been shown in three different graphs (Figures 4,6.1, 4.6.2
and 4.6.3) on the ordinate (y-axis).
Levels of Hierarchy; It is clearly visible on the scatter
diagrams (Figures 4.6.1, 4.6.2 and 4.6.3), that their occur
four distinct breaks in social and demographic variables.
Three breaks have found in total composite index of progression
of the slope resulting in the separation of five and four
separate grades or classes. In 4.6.1 diagram first and second
breaks are more conspicuous while the third and fourth breaks
are less obvious. A close scrutiny of appendix-4 reveals the
fact more explicity. Thus, all the 75 urban centers (SMA and
Municipality) in Bangladesh have been categorized into five
different orders on the basis of social facilities and demogra
phic indicators (Figure 4.6.2) but four different order can
be shown in Figure 4.6.3 of overall facilities of urban centers,
Maps 4.6.1, 4.6.2, and 4.6.3 illustrate the spatial distribu
tion of different order of urban centers in Bangladesh. These
orders are indicating the different hierarchical classes.
According to an importance of the urban center a factor of
urban dynamics and the impact of demographic correlates on
urban system changes in different phases have been formulated.
For the further analysis approach can be adopted.
-w^
CD CD Oi
X LU Q
LU
en o Q.
o o
<
if) CO
< _ I o
CO J "
I
UJ cc ID CD LE
o
o z < liJ a: m
0 1
V) tn en J-
i/i a< 1 -o u w a> Ol
IN
l/l a> L. o o in lO
03 ^
0 3 CO
c
a< ICC
I I 1 V I 1 ! •
ID
CJi C 0 Ol 10
o
Q LU tn < CD
oo Q: LU h-z LU o z < CO ce ZD
^, trt Ol
03
< 5 <
to
i^ z <. cr
*" o z <
• - II
z <
II
o ^ ' - Z
<
II :«: z <
c
3 x:
1-1
a ' Ol I c
•i .»-.;;
o to
o o J-
o o CO
o o CM
< a.
3 Z
<
a: LU
CO m
cvJ
o o
i N 3 0 a 3 d NI X3aNI 3ilS0dW03
136
c •H o 0.
01 c
•H J: '0 01
u
(0 <u M O u
\0
in 0) M o u en
U) 1) u 0 o to
Ul (U u 0 o m
o
It]
a % % \ -
>Ji >
en c o Ol 10
•H
u
crt O Q. S O o z o o lU Ul
< rD W) Q: Hi y~
z lU o z < 00 a: 3
in td >j CQ < OS •t
> u M s cu $ t3 o Z bl
o
c
>4
C 10 IE
C a
X K
< — U M
M
z o M
H 3 a £
c
O w
CVI
J-
LU
a: O
J3 10
x: <a a:
c o
c o —
O "0 £ -J u •
03 r-( •
C
O
as Cd m x:
' ^ V :
o ^4 at
o o OV
o r* 33
o ^ CO
o ^H CO
o CO c
o i n r-
o <N r-
o o o o o o o o o o o o o o o o
xN30Had Ni xaoNi axisodwoo
o o o o o o •-H 00 in (N ON \o rvl ^ ^ rH
TJT
o a.
o z < UJ
m
in CM
J-
^ B
(±T-IT)
s
-o o I
03
X tiJ Q
LU
in o Q.
o
-o
Q LU (/) < CD
cn CY. LU 1— z LU C )
^ < m a: ZD
LL O
w UJ _i m < oc < > _ i
u o (/) -
^
<
CO CO < - J o CO
(£J
• LU CC ZD O
LL
i n
3 C
lf> U
z <
<
2 >
to
z <
o
o»
CO
r-
U9
u>
J-
CO
I :<: <
z
z o ~^ <
J
a. o a
z < K
^ II
z < ex
^ z
>-_ l
CJ
z 73
O z < <
u. o
cc LU CD
z
iNBoaad Ni x3aNi aiisocjwoo
138
<^> BANGLADESH
.^••^ A SPATIAL DISTRIBUTION OF FIRST, SECOND, THIRD r" \^"'.::,< ORDER FUNCTION ( if DEMOGRAPH IC VARIABLES )
I.
i- r
\. ' • * )
\
A
Q 20 ^0 6,0 80 100
Kilometers
\ j s
,..>
•p
r
o KHULNA
Nr\/
o DHAKA
C
\
2VN-
p.,yv:
0 • • . • • • •
r S
INDEX
O First order City
O Second order City
o Third order City
4- JL
R e i u a n / ' 8 8 Map-if.6.1
139
C: BANGLADESH
^••\^..^ X*.. SPATIAL DISTRIBUTION OF FIRST SECOND AND THIRD y " " \^;':r^\ ORDER FUNCTION( 11 SOCIAL VARIABLES )
\ .
\ . A . V S
o Rangpur
\ 0 20 t+O 60 80 100 I I I I I I
Kilometers , . .A
rK.f O Bogra
Sherpur
^ . • V , " • - v . .
s. « ..J
I (
J
•\ •0
• - ^
O Tangai l
O DHAKA
^^•J r -^
iww\
c 0- /A
\\ ' W \
Chittagong
i'yjS/^ NDEX
O First order City
O Second order City
o Third order Cities I 9igiE
)
:-" v . . V
Reiuan/'88 Map-£+.6.2
140
BANGLADESH
A. SPATIAL DISTRIBUTION OF FIRST SECOND AND
./•''' \ ^ S \ THIRD ORDER FUNCTION (15 VARIABLES) I • •"V
_ / '"^"vj
O Rangpur
I v..
(
i r-. .~
/.
v....
0 20 kO 60 80 100 1 I I r . I i
Kilometers
/
I
^ -
,..-'
^ ) " ^ . i
2k N-"
o Khulna
O DHAKA r \ \
J 1
• J •. .-> V . . ' ' '
r
NDEX O First order City
O Second order City
o Third order Cities
9,0°E
Chi t t agong
Reiuan/'88
Map-^.6.3
141
Pattern of classification of settlement; The accompanying
illustrations (Figures 4.6.1 to 4.6.3) and table 4.6.2 bring
out the salient features of the arrangement of urban settlements
into at least five classes (groups). The classification is
based on the gaps in composite score. The frequency distribution
of urban settlements also brings out the underlying pattern
of relationship between population size of the urban centers
and the composite rank. The following features are noteworthy:
i) As could be expected, Dhaka SMA stands out as the
first ranking urban center and functional gap (as
determined by the difference in composite score) between
Dhaka SMA and the second order urban center is very
wide in all the three alternative schemes of ranking
(see figures and table).
All the three type of classification about double composite
score have been obtained by first rank order of Dhaka
SMA. Dhaka SMA scored the top position due to being
National Capital and an industrial city. On a per
capita composite score Dhaka SMA obtained only 0.002
but the Ishardi (last rank) obtained 0.01. So it is
notable that the percipata facilities in lower order
urban centers are better than in the higher order.
ii) Chittagong SMA is in second rank order, having the
composite score of 204, 475, and 277 per cent is social,
Demographic and over all weightage. Chittagong obtained
the second position due to being sea port city about
75 per cent of export and import of goods is through
the Chittagong port. Due to that most of the large
scale and heavy industries developed in the Chittagong
SMA area. Khulna and Rajshahi SMA became 3rd and fourth
in Demographic composite index but these are 8th and
13th in social facilities and 3rd and 9th in over all
142
weightage score respectively. In over all weightage
Khulna became 3rd because it is the 2nd largest sea
port of the country. There are Food and different
pulp and paper, and Board mills in and around the
Khulna SMA. But as a Metropolitan city Rajshahi could
not achive the 4th position with relation to popula
tion. The social facilities are not optimum with
relation to other SMA. Whereas Rangpur because 4th
in position in overall score due to sugar mills and
old administrative activity. But on the other hand
five district headquarters (except Sherpur) achived
3rd rank in social and administrative facilities.
Out of 7 5 SMA and Municipalities only 4 SMA have been
engoying 76.7 per cent of social facilities of the
country.
iii) There are several instances where the urban settlements
are large enough in terms of population and yet func
tionally their rank is low. e.g. Mymensingh, Saidpur,
Serajganj etc. and this is because of absence of good
road communication.
iv) Urban centers ranked on the basis of social facilities
and amunities show that a relatively large number
of settlement are in lower hierarchy (hierarchy number 5)
of settlements (57 in number).
v) On the whole the structure of urban settlement ranking
brings out a high degree of imbalance with very few
urban centers acquiring services, facilities and func
tions to the optimum level. Demographic correlates
are affecting some urban places like Rangamadi Chowmohoni
(Begamganj), Jhenaidah etc. The impact of demographic
correlates on urban system, would be observed in corre
lation matrix.
143
Urban settlement hierarchy: The regional setting, the urban
settlement structure and spatial pattern of distribution and
the ranking of urban settlements provide some clues to the
existing pattern of urban settlement hierarchy and space rela
tions. Just as a large size of urban center need not necessarily
have a higher rank on the basis of composite functions, all
urban settlements of higher rank might not attain a higher
level in the settlement hierarchy. Here the concept of hierarchy
is a factor of urban dynamics; and different levels of urban
hierarchy would reflect different levels of .nodality expressed
through the convergence of people and goods in different places.
Analysis of correlation Matrix; From the total 25 correlates
a small set of 11 veriables were chosen for the measurement
of association and ranking of urban centers after the use
of composite index.
The salient features of the correlation matrix (see table
4.6.3) are as follows:
i) 11 X 11 correlation matrix of 75 SMA and Municipalities
of Bangladesh indicates some high inter correlation,
specially among social facilities and amenities. But
in some cases demographic correlates also gives a
better relationship with the social facilities. That
is demographic variables also contributes a good amount
of weights in urban system development. It was observed
that population size, services and facilities have
a very high degree of correlation (about positive
0.99 ). This brings out the facts that a certain level
of population is needed in order to provide services
and facilities of different types and levels. It was
also confirmed by the weightage assigned to different
services and facilities on basis of their frequency
144
of distribution among different size range of urban settle
ments. While population size is important in providing services
and facilities there is a need for a rational policy for their
spatial distribution because in the Bangladesh context number
of urban centers of.different size classes at any time, requiring
these facilities, would far exceed the actual provision in
the plan allocation. There is also spatial irregularity in
the distribution of the urban settlements of different sizes.
It is also important to note that the association of population
size with non-agricultural activities is rather weak but is
positive (0.69). The non-agricultural population and the
services and facilities also show slightly weak correlations.
The values of correlation are 0.65, 0.57, 0.71, 0.37, 0.58
and 0.67 for the physician, pacca road. Bank, telephone. Medical
facilities, student and educational facilities respectively.
ii)Physician, Bank, Telephone, Medical facilities and education
show a very high degree of association among them. Their intera
ction values lie above + 0.88.
Lii)Some variables are highly correlated among them but they have
no direct relationship but through indirect urban dynamics
factors which have influenced a caused high degree of correla
tion, like-literacy rate and pacca road. Non-agricultural popu
lation and pacca road. Physician and Bank, Physician and student.
Physician and educational Institution, Pacca raod and Medical
facilities, telephone and student enrolment and educational
Institution, etc.
iv)The table also indicated that some variables have high positive
correlation but in actual field the data have shown low positive
correlation like-literary rate and Bank, student enrolement,
educational institution, urban growth rate and Pacca road,
urban growth rate and non-agricultural population etc. But
the values is less than + 0.65. This is because of the low
levels of services and facilities and economic activities in
the most small urban centers. So in some cases urban dynamics
145
variables became very weak due to unplanned urban development.
The input is irrational due to government policy which favours
the elite group of peoples.
v) With the increase in population size there is a natural tendency
for diversification of activities and services and facilities
leading to specialization of each function at still higher
levels of urban development.
So, the correlation matrix has given some indication of
relationships of different correlates and influence of demogra
phic correlates in urban development programme. Composite in
dex and correlation matrix both are the closely related and
gives the results of urban dynamics trend and national develop
ment strategy for better future of urban expansion and new
urban growth policy.
146 00 00 ON
en
o •H T3 C
• H
(I) • P •H (0 O
a E O o 0
CO
• H w (0 J3
0)
C o (0 iH 0) +J c (U u c to
X3 V
D
it-i O
c o
• H +J (d u
• H M-l • H U] (0 (0 H U
x: (0
•o (0
H
en c (0
CQ
0) H .O rtJ
E l
_ > u
• — •
CO • M
x; en
• H
<U 3:
m 0)
H XI (0
• H l (0 > 0
•H
x: a (0 M 0 1 0 E <u Q
T
U] (1)
H X I (0
- H 1 Id >
H 10
• H 0 0
C/J
i-H
i H
•- U) m (0
H U
1
1 c
S-l 0)
dj a - ^ Di <U c c ? (0 -H "d K — + J
1 3 cr >i (U O M C b^ (D
1 c 0) u M (U
•i' "^ T CJ) 0 c c 5* (0 -H "d
OS ^ + J
1 3 CT >i (U O M C b 0)
1 1
c d) u u 0)
<D a, "^ 0 1 • Si
c c H' fO - H (d OS -^-^^
a a 0 M 0 1
t-H
o O n A
^
o o m / \
.—(
i n ( N <N
A
M
r-t
o o m
I o o CN
r - l
o o i n
1 o o n
i-H
i n ( N ( N
1 o i n r H
H H
( N
O O f N
1 O r-i H
r H
o o n 1
o r~-(\
vo
o in r H
1 o o r H
M M M
i H
O r-t H
V
( N
O r~ fSl
1 i H
i n r-i
t H r H
O O i H
1 r-t ^
> M
1
1
O r-
r H
i n <-i
V
r i n
r~ r-
V
>
in
1
in r>-
1
in r-
• 1
i H
to +-I o E-1
to 0
•H • M
(0 • H -M (d
- p CO
> 1 +J • H H (d ft
• H O •H C 3 s:
s M
z r 00 <y> r H
. (d (U M (d
c (d
3
C o
0 ft
IX,
•a c (d
CO (U CO c (U u
4-1
0 •H
^-| CO
•w Q
. . r-
in 00 CQ O N
CQ - 1
„ QJ 0 -1
3 0
CO
• CO
(1) H XI (d
+ j
0) M 0 0 CO
d) • M
• H CO 0 ft E 0 u
(U x: • M
E 0 >^
o J= -t-J
3 (d
<u x; - p
>-X3
n QJ
• p
(d
3 U
H (d u
(d + j
to Q *
u u o u
H O C/3 2
n
>
ON
fN
- n : t T : ; s u i [BHOTItEOnpa
^uauifojua s:;uapn4S
XBOxpsw
3uoqdsx3i ,
5(UBa
peoH BOOBd
UBTDTSAqj
uoxq.
n
147
i n
-BX^ltiOd '6v-uoN
udod uBqan
84.BJ Aoej9q.T7
l-t
•1
m •
VO
• ^ 0) H X3 (0 EH
• • H 0 00 2
H X) (D •H M (0 >
IT) «4
•^ T
o
o in
o i-H
o
o
in
o
O
O O
(a u
u
i n
00 VO
i n
o
o
fN in
o o
Q) -M (0
o en
c (0
D
• ^
m fN (Ti
00 00
VO
(Ti
(N as <Ti
i n
o
o o
in VO
00 o o
VO
O O
O .H
O O
a o a,
en < c o 2
n
c o
•H
3 a o a
(0 -M
o
(N
c (0
•H
u • H U] > i
-a ID o M
(a o o (0
a,
a>
X c to
CQ
C
o
a 0)
o EH
VO
O O
o
no VO
o
^ i-H
o
r-VO
o
o r-
o
iH
1^
o
r-VO
o
a\ a\
o
<Ti as
o
as as
o
as as
o
in as
o
^ <Ti
O
as as
o
o o
o
o o 1—1
rH
(0 0)
•H O (0
4-1
(0 U
•H
n
as
c 0) e
H O U
c 0)
c o
-M C/5
ro
O
o o
(N
^
O
VO VO
o
00 in
o
as as
o
as as
o
o as
o
r 00
o
V as
o
(Ti (Ti
o
o o
I-H
c o
-H
c M
C
o •H +J (0
u X3 W
m
vO
CTi
rsj
n
0) H XI
•H
M
> 4-1 O
^1 QJ j a E D C
148
4.7 RANK SIZE RULE;
After analizing the hierarchy of urban centers, another
hierarchical concept is used which considers the size of cities
as well as their spacing, in relation to population size of the
cities or towns. This concept is popularly known as "rank-size
rule". Although in 1913 Felix Auerbach had indicated that there
is some regular relationship between rank and population size
of an urban settlement, the concept was systematised by George
K. Zipf (1958, pp. 83-91) through his empirical findings. Hence,
this rule is also known as 'Zipf's rule'. This rule states that
all the towns in a region are arranged in a descending order
according to their size of population (The largest urban field
is given rank 1) and plotted in a graph which has the logarithm
of population on the ordinate (y-axis) and logarithm of rank
on the abscissa (x-axis) (Berry and Horton, 1970, p. 69). Basically,
the population of the second-largest city will have half the
population of the largest or primate city; the third-largest city
will have one-third of the population of the largest city; the
fourth largest city will have a quarter the population of the
largest city; and so forth. However, the size of the population
of n town will be 1/n the population size of the largest city,
and thus the ordering arrangement of population size of the other
urban settlements may be worked out according to a series in
the form of 1, 1/2, 1/3, 1/4, 1/n The relationship between
size and the rank and may be expressed statistically in the follo
wing formula ( Hartshorn 1980):
P = P. . R "•'• n I n
where P = population of the city to be calculated n P, = population of largest city
R = rank of city to be calculated n
For example, one could calculate the size of the third largest
city in an area as follows (assuming the largest city had a popu
lation of 17,000,000):
149
P3 = P^. 3 -1
P3 = 17,000,000 X 1/3
P3 = 5,666.667
This calculation indicates that the third-ranking city would
have a population of 5,666,667 persons.
Graphing the rank and size of _a group of cities produces
curvilinear (Logarithmic scale) relationship (what the statisti
cians call a "reverse-J" frequency distribution) as indicated
in figure 5.7.1 and 5.7.2 shows the straight line relationship
(arithmatic scale) of rank-size rule.
RANK-SIZE ROLE 5-
4 .
^ 3 -c 0 •rH -M
^ 2-0 Cu S 1-
0
r- 1 \ \
V
\ \' \ \ \ \ o\
;
\ \ \ s
N c
2
V *v
"!
3 4
0) 4 -
ta u ^ w 3^ en 52. c 0
•H
tl 1-r-i 3 a s.
c
< \ \ \ \ \ \ \ \ \
\
1 :
\ \ \ \ \ <
I
\^ ^ >, t
3 Fig.5.7.1 Arithmetic scale Fig.5.7.2 Logarithmic scale
Some of the authors, like steward, in his work, reviewed the
rank-size rule and examined some, determinants of towns spacing.
He arrived at the conclusion that the rank-size rule has no logical
basis but breaks down in many areas. He also stated that there
is nothing natural about the shape of the town-size pyramid but
that relative population sizes of urban settlements in different
functional classes and their relative number vary with the techno
logical stage and economic development of the area. Berry and
Garrison while discussing rank-size relationship even question
the very idea of regularity? A scatter of dots formed by plotting
the size of cities versus their rank never presents a perfect
alignments of dots along a mean line. One dot will fall above
the line, another below, and so on (Berry and Garrison, 1958,
pp. 83-91).
150
Application of the Rank-Size Rule; Bangladesh; An attempt has
been made to test the rank-size rate in Bangladesh to find out
rank and population size relationship and in what way this rule
has similarity with hierarchy of urban centers of Bangladesh.
According to 1981 census, total urban population is
13,536,000 i.e., about 16 per cent of total population which
comprises 492 urban centers, but for the test of the applicability
of the rank-size rule, only 13 major urban centers have been
chosen having more the one lakh population. The range of popula
tion is from 3,440,000 (Dhaka) to 107,000 (Sirajganj). To test
the applicability of the rule first of all urban settlements
of the area have been set on a graph in the way that the population
size of cities comes on ordinate (y-axis) while the rank of towns
in descending order are put on abscissa (x-axis). Figure 4.7.3
shows that the curve thus obtained does not reveal any linear
relationship, as it should emerge out in ideal conditions given
by Zipf. A straight line has been drawn to join the dots on the
scale, but it is not in conformity with the rank-size rule. It
has no resemblance with the intermediate 'distribution type' as
discussed by Berry in his work (1961, pp. 575-88). Position of
primary to top city would be reflected on the graph 4.7 J by a
large gap between the points representing city of largest popula
tion, i.e., Dhaka, ranked number one, and that for the second
largest city, Chittagong. A large gap in the hierarchy of popula
tion size becomes visible. Proportions of population sizes of
the towns considerably eleviate from the ideal pattern as given
in the concept. The position of the points indicating respective
centers has a wide gap from straight line. This suggests how much
the urban centers of Bangladesh experience an irregular size of
population in the progression. However, in ideal conditions, this
relationship appears as a straight line. In a general sense the
rank-size rule is a common measure of urban status of a nation.
151
W
C5
CQ
QS-
u 2 U u 2 < QQ Q:
Q: O
< 2
X
z o
UJ OC
111 N
<n I
z <
o
m < o o < cr Q.
B5{Bqa
o o n o
v J L
in
ro
(N
(Ti
00
r
vO
U t4 M
cn X OS
w u M H M
u b O X, z < K
- in
- m
oo 00 VO
HMVT NI NOIiVindOd
152
Table 4.7.1 represents the actual and expected population
size of the cities. The figures show a very poor fit, and the
reason for this is the extensive size of Dhaka Statistical
Metropolitan Area (SMA) which is national capital of Bangladesh.
The other cities have become so dominant that its relative growth
has been much greater than of other cities of the country. It
tends to attract high quality resources in all respects, both
from rural as well as from other urban areas. Only Chittagong
and Khulna show the minimum percentage of deviation from Dhaka
SMA. The table also indicated that the higher the order of urban
centers on the scale of population hierarchy lesser is the per
centage of deviation, and smaller the size of population more
or less of the urban centers, higher the percentage of deviation.
It is need that the degree of concentration of centralized
services and the population size of respective centers would
have been evenly developed, the degree of divergence in the
hierarchical arrangement of urban centers as well as in the
rank-size relationship would be minimized.
Moreover, the table clearly focuses the fact that all
the cities (urban centers) except the largest one, in the study
area have actual sizes lesser than the expected size. For the
dominance of Dhaka SMA all the other urban centers disturbed
their actual growth. So, the growth of larger urban centers
are faster and small urban centers are slow growth. The largest
city influences the entire country (territory) and it is proved
in the present case. The primacy of Dhaka is established beyond
doubt. It has much greater population than normally expected,
exceeding 19 lakh in 1974 and 35 lakh in 1981 census. Over the
decade several extra function such as beg air port. New Islamic
University, different types of new industries were added a large
population for extra-facilities and job was added.
c (U o u a> cu ( 0
ro 0) 0 C 0) Ul (U iw M-l
•H Q
H (0 3
• M
o to
c (1) C)
s • p
(1)
Q) N
• H
CO
H (0 3
+ J
u (t3
m 0
0) N
• H
U)
J3 TS
(1) O c 0) M <U
M-l (4-1
• H
Q
-o <U
+ J
O (U
(U • P
o (U a X (U
T3 C (0
c 0
• H + J
10 H
D a a X u
H (0 0
• M
o <
H (0 U O U
0 cu c 0 •H • M
(0 H D a 0
O i
m a 0 •H u 0) a:
^ c (0 ce
• H 4-1
(0 Q:
(0 H
iw 3 C 0 a 0
0 -H Oi -M
(0 0)
•H +J •H U
153
<Ti
ȣ)
n rvj
o r~
i n
r-
VO
a\ CO
n CN
CT> M
o > CN
^ r H
O . - 1
f N
CN i n
o CO t-i
o 'a* i n i n t-H
o n
<Ti
^ i-H
o o i H
ro :—(
• i n
^ "S* <-{
o CO
I—I VO
r H
VD
0 0
' J '
"a* ^
o as n
<n ( N
n
ON VO
i n
m <Ti
^
r H
• H
n VX3
o VD
0 0
i H
< H
r-CTi
^
rvj m o>
VO
CO
m
vO
o^ i n
i n r H
n
r ^ VO
r H
VO
O J
O J
CO
VO
0 0
f N ( N
CO
CO
O
i n
c r i
r H
i n o o n 0 0
r H
r-~ vo •^ VO
r>-r H
r~-r-~ VO
T
m r H
r T r H
o • T
^ r-o
o rsi [ ^
( j i VO
i n
i n
• V r H
r~ n o o VO
CO
<Ti CN
o CO
CO
VO
T VO
O
r H
1 ^
i n
r H
o i n
O i
CO
T
CO r H
o o n •
VO
i n
0 0
i H
CO
m
i n r H
o T
• V
n
n r H
VO
<Ti
o n
( M
ro i n
i n 0 0
CN
• H
i n
•«r
i H
VO
( N
n
r-' T r H
« o ^ ^
« n
• ^
CO
VO
> o a\ n
> r H
O
o o ( N
i n
VO
VO
CN
r
n i n CN
r H
i H
<Ti
o CTi r H
O J
en r H
^ 0 0 r H
i n
o CTi
CN
r-i H
r H
r-n CO
VO
i H
•<T
r-r H
n i n i H
r-CN CTi
CO
^ r H
CO
o VO
vO
CN . H
i n
VO
o <Tv
o r H
• r-r-» VO
o i H
o o o
o o i n
r") ro m
o i n
CN
o o CM
VO
VO r H
CN
• r H
i n CN r H
CM
o o r H
o CTi
O
ro 0 0
o
VO
r-O
ro in VO 00 as CN ro
03 X ro
£ :
en c 0 en (0
• p
- p
• H
£
(0 c
r H
3 x:
• H
x; (0 £1 ( 0
• r o
(0
en c
• H
M
C (U E > i
(0 H H •H
e 0
H
(0 in •H V4 <0
p
0) x: H > i
u 3 a 0 1
c (0
0) ^ 0 m ( 0
OJ
M 3 a 73 • H
(0
03 C X) 03
• n C 03 Cn
•r- i (0 l-l
• H
0)
w 03 (U M U
c •H
> 03
rH 3 O
•H
u x: u 03 0)
x: u 3 E
o x:
p 03 U
•H TD C
C cn •H
3
0) p 0 Z
CJ o: u m w K C/3 U3
CHAPTER - V
CONCLUSION
The purpose of the preset investigation was to get more
insight into the mechanisms of the urban dynamics in the case
of Bangladesh. Impact of demographic correlates on urban growth
in relation to spatial, functional and population aspect. Before
the concluding remark and suggestion or recommendation of the
research paper, a brief findings of the different chapters can
be pointed out.
The urban dynamics study of the towns and cities was mainly
made in the western world, but its application is rare in the
cities of the developing and underdeveloped countries. The main
purpose of the present study was to identify the basic factors
of urban dynamics and specially the impact of demographic correla
tes on urban system change of all four statistical metropolitan
areas and 71 municipal towns and cities of Bangladesh.
Aims, objectives and methodology of the study has been
illustrated in the first chapter in which a systematic overview
of spatial and function changes of urban area of Bangladesh have
also been studied.
Urban geographers have made a more substantial contribution
than population geographers in urban dynamics study of the world
level and have tried to identify the nature and characteristics
of urban dynamics in near future.
Historical and geographical factors of an area can be
focused through analyse the future level and trend of urban system
of Bangaldesh. Chapter three has investigated the historical
and geographical implication of spatio-temporal factors. It is
evident that a greater part of the riverine and wet Bengal delta
and the inland low plains were under forest and/or marsh land
155
through the pre-historic and even in the historical past.
Even as late as in the Mugal period, much of the Gangetic
plain was under forest and human occupations was quite slow
to penetrate into such lands. There is evidence that major
pockets of human settlements existed in fertile river valleys
of agricultural value. One the other hand, evidence from popu
lation history of this region suggests that very few and small
population settlement developed due to number of natural
hazards, like-flood, storm surges, sudden shift in river course
etc. which undoubtedly led to the decline of flourishing human
settlements, specially in the southern part of Bangladesh.
On the other hand, it led to the migration of the rural people
to urban area and lead to urban population and territorial
expantion. The people are migrating from southern region to
central region and the people of Noakhali moved towards
Chittagong urban area. Moreover, river acts as an important
controlling factor for the origin and development of urban
centers. It can be mentioned that the world ancient cities
have originated near the riverine area and the Bangladesh
is not exception to this phenomenon.
In present time, all the big cities have developed near
the riverine area, like-Dhaka, Chittagong, Khulna, Rajshahi,
Mymensingh, Barisal, Sylhet etc. in the bank of Buriganga,
Karnafuli, Rupsha, Padma, Old Bramaputra, Lower Meghna and
Surma rivers respectably. High attitude and dense forest,
and unfertile lands are inversely related to the urban center
growth and development. Due to high altitude and dense forest
of Chittagong hill tracts there are very few urban centers.
Modhupur tract and Sunderban area are also discourage urban
growths due to dense forest and water. Barren tract in northern
part of Bangladesh has sparsely developed urban centers due
to unfertile piedmont region.
156
Apart from agricultural aspects physical factors play
a vital role in urban growth and development because it brings
out better differences in linkage facilities. Due to the low
road and railway connectivity - Chittagong hill tracts, southern
part of Khulna, Rajshahi division and north-eastern part of
Bangladesh have fewer number of urban centers. On the other
hand, Dhaka, Chittagong, Comilla, Mymensingh, Tangail, Jamalpur
districts have high density of urban centers, only because
of better road and railway connection. This high connectivity
is also because of its favourable physical environment.
The main focus and finding of the chapter four is the
identification of urban dynamics variables through different
statistical techniques-like spatio-tempural analysis, hierar
chical classification through composite index, and rank-size
rule of the major urban areas.
From 1951 to 1974 period the overall pattern of urban
spacing was random but after 1981 the pattern became approach
ing to uniform due to decentralization of administration.
Urban hierarchy was also computed through composite index
and that shows that Dhaka SMA, Chittagong SMA and Khulna
SMA secured the 1st, 2nd and 3rd rank but Rajshahi SMA has
fail to scored the 4th position. Correlation matrix shows
a very high inter-correlation between demographic and urban
facilities. Rank-size Rule has failed to prove the general
progression of the 13 major cities.
In order to identify the strong correlates of urban system
change, different statistical method and techniques have been
applied. Only 15 variables have been selected for measuring
the urban hierarchy but this small number of variables is
rather inadequate to identify the real position of urban dyna
mics machanisms, the study needs more variables, specially
the demographic correlates. For better measurement of urban
157
dynamics - a set of socio-economic and other relevent variables
are to be included in Ph.D. research. In further study the
principal component analysis might be used for surching the
first principal component that explain the real urban growth
in relation to population, function and spatial expansion
of the study area.
158
Appendix-1: Percent and Rank of Urban Population by District in
census years 1961-1981
Country/ Districts
Bangladesh
Dhaka
Chittagong
Chittagong H.T.
Khulna
Bandarban
Kushtia
Barisal
Pabna
Noakhali
Rangpur
Jessore
Rajshahi
Mymensingh
Jamalpur
Patuakhali
Dinajpur
Sylhet
Faridpur
Comilla
Tangail
Bagra
Percent
5.19
14.79
12.49
, 5.92
7.04
-
5.42
3.49
5.08
1.44
4.20
3.44
2.28
1.11
4.38
1.03
4.21
2.03
2.47
3.17
1.59
2.98
1961
Rank
-
1
2
4
3
-
5
10
6
18
9
11
15
19
7
20
8
16
14
12
17
13
1974
Percent
8.78
29.57
20.98
9.78
14.63
11.49
8.30
3.93
7.62
2.15
4.82
5.43
5,78
5.94
4.64
2.50
4.41
2.77
2.87
4.24
5.24
3.71
Rank
-
1
2
5
3
4
6
16
7
21
12
10
9
8
13
20
14
19
18
15
11
17
1981
Percent
15.54
38.86
32.10
30.94
22.51
21.56
14.10
12.10
11.81
11.30
11.29
10.96
10.85
10.24
9.10
8.99
8.87
8.74
8.58
8.51
7.56
7.44
Rank
-
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Source: BBS, 1981 census, p. 13
159
i^pp«idix-2: Dis t r ibut ion of Urban Population by Di s t r i c t in Census year 1961-81.
Division/ District
Bangladesh
Ctg.Division
Bandarban
Ctg.H.T.
Chittagong
Comilla
Noakhali
Sylhet
1961
Number* Per-
2641
640
-
23
373
139
34
71
Dhaka Division 1073
Dhaka
Faridpur
Jamalpur
Mymensingh
Tangail
Khulna Division
Earisal
Jessore
Khulna
Kushtia
Patuakhali
Rajshahi Division
Bogra
Dinajpur
Pabna
Rajshahi
Rangpur
754
78
63
154
24
430
107
75
173
63
12
498
47
72
100
120
159
cent
100.0
24.2
-
0.9
14.1
5.2.
1.3
2.7
40.6
23.5
3.0
2.4
5.8
0.9
16.3
4.1
2.8
6.5
2.4
0.5
18.9
1.8
2.7
3.8
4.6
6.0
Number
6274
1405
13
38
906
247
69
132
2900
2250
117
96
328
109
1049
154
181
520
157
37
920
83
113
215
242
262
1974
•* Percent
100.0
22.4
0.2
0.6
14.4
4.0
1.1
2.1
46.2
35.9
1.9
1.5
5.2
1.7-
16.7
2.4
2.9
8.3
2.5
0.6
14.7
1.3
1.8
3.4
4.0
4.2
1981
Number*
13536
3487
37
179
1762
586
430
493
5383
3900
408
113
667
185
2468
565
441
974
323
165
2198
203
284
404
572
735
Percent
100.0
25.8
0.3
1.3
13.0
4.3
3.2
3.7
39.8
28.8
3.0
1.7
4.9
1.4
18 i 2
4.2
3.3
7.2
2.3
1.2
16.2
1.5
2.1
3.0
4.2
5.4
Percen
1974-74
138
120
-
68
143
77
103
86
170
199
51
52
113
360
144
44
140
202
147
204
85
76
58
116
105
65
t increaseih
1974-81
116
149
185
374
95
137
529
274
85
73
249
132
105
7 0
161
267
143
88
106
349
139
145
251
188
132
181
1961-81
412
446
-
683
373
322
1170
594
401
415
424
253
438
671
474
428
488
463
413
1283
341
332
294
304
377
362
*Numbers a r e t h o u s a n d ; S o u r c e : BBS, 1981 c e n s u s .
160
Appendix 3: Bangladesh: Growth of Urban Population 1951-1961 and 1974-1981
1961-1974
(Data by District)
Division/ District
1951-1961 1961-1974
increased in increased in
1974-1981
increased in
Number Percent Number Percent Number Percent
Bangladesh
Chittagong Division
Bandarban
8,20,953
1,44,414
R Ctg. Hill Tract R Chittagong Comilla Noakhali Sylhet
76,778 22,353 12,650 9,829
Dhaka Division4,22,007
Dhaka Faridpur Jamalpur Mymensingh Tangail
Khulna Division
Barisal Jessorc Khulna Kushtia Patuakhali
Rajshahi Division
Bogra Dinajpur Pabna Rojshahi Rangpur
2,20,784 20,931 16,698 39,777 2,175
1,51,079
- 16,291 37,898
1,03,461 22,730 3,283
1,03,453
10,551 - 4,592 30,041 35,634 31,819
45.1
16.1
R R
25.6 19.2 58.6 16.1
64.8
65.7 36.3 36.1 34.9 10.1
54.1
-15.8 101.3 150.0 56.1 13.2
26.2
+ 29.0 - 6.0 43.2 42.1 24.9
3,633,000
765,000
R 15,000
533,000 108,000 35,000 61,000
1,827,000
1,496,000 39,000 33,000 174,000 85,000
619,000
47,000 106,000 347,000 94,000 25,000
422,000
36,000 41,000 115,000 127,000 103,000
138
120
R 68
143 77
103 86
170
199 51 52
113 360
144
44 140 202 147 204
85
76 58
116 105 65
7261
2082
24 141 856 339 361 361
2483
1650 291 127 339 76
1475
411 260 454 166 126
1276
120 171 185 325 473
000
000
000 000 000 000 000 000
000
000 000 000 000 000
000
000 000 000 000 000
000
000 000 000 000 000
116
149
184 374 95 137 529 274
85
73 249 132 105 70
161
267 143 88 106 349
139
145 251 188 132 181
Source: BBS, Census, 1951, urban census 1987
1961, 1974 and 1981 national volume and
R = Rural character
161
Appendix 4: Ranking of Bangadesh Urban Centers (SMA and Municipality)
on the basis of Hierarchical concept.
SI. No.
1
Name of the Municipality /SMA
2
Composite index using social facilities per lakh population on a ranking criteria
3
Indices
4
Rank
5
Composite Indices index using demographic variables per lakh population as a ranking criteria
6 7
Rank
8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16.
17
18
19
20
21
22
23
24
25
Dhaka
Chittagong
Khulna
Rajshahi
Cos's Bazar
Rangamati
B. Bria
Chandpur
Comilla
Feni
Laxmipur
Chowmuhani (Begangong)
Noakhali (Sudharam)
Hobiganj
Maulavi Bazar
Srimangal
Sunamganj
Sylhet
Manikganj
Munshiganj
Norsingdi
Faridpur
Rajbari
Gopalgonj
Madaripur
2691.2
2247.3
1617.9
1362.0
1078.1
865.1
922.1
892.6
1428.5
1025.9
600.8
519.1
1077.9
701.8
1048,1
895.0
632.9
1417.5
799.6
1023.1
988.7
1181.5
749.8
850.4
643.2
244.65
204.3
147.08
123.82
98.00
78.65
83.83
81.15
129.86
93.26
54.62
47.19
97.99
63.80
95.28
81.36
57.54
128.56
72.69
93.01
89.88
107.41
68.16
77.3
58.5
1
2
8
13
22
41
35
40
10
25
71
74
23
58
24
39
68
11
50
26
27
18 •
54
43
66
3637.4
1901.4
1091.7
610.7
342.1
508.2
290.6
354.8
421.5
341.2
151.3
610.0
312.1
309.2
318.5
344.8
291.2
418.4
323.9
410.6
441.4
344.7
265.3
288.3
253.2 Coi
909.4
475.4
272.9
152.7
85.5
127.1
72.7
88.7
105.4
85.3
37.8
150.5
78.0
77.3
79.6
86.2
72.8
104.6
81.0
102.7
110.4
86.2
66.3
72.1
63.3
1
2
3
4
31
6
49
28
13
32
73
5
41
42
38
29
98
14
36
17
10
30
58
51
62 ntd.... 2
162
1
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
2
Jamalpur
Sherpur
Bajitpur
Bhairab
Kishorgonj
Gouripur
Mymensingh
Muktagacha
Mohongonj
Netrokona
Gopalpur
Tangail
Barisal
Bhola
Jhalokati
Pirojpur
Jessore
Jhenaidah
Kotchandpur
Maheshpur
Magura
Kalia
Narail
Bagarhat
Monglaport
{Rampal)
Satkhira
Chuadanga
Kumarkhali
Kushtia
Meherpur
3
782.4
691.2
905.1
569.7
933.4
1093.3
1168.1
918.4
1090.2
826.1
914.2
1652.7
1896.8
722.6
841.8
944.5
1349.8
957.8
931.3
802.9
262.5
798.5
859.0
693.5
609.8
667.0
733.3
965.2
1477.5
932.8
4
71.1
62.8
82.3
51.8
84.9
99.4
106.2
83.5
99.2
75.1
83.1
150.3
172.4
65.7
76.5
85.9
122.7
87.1
84.7
73.0
114.8
72.6
78.1
63.1
55.4
60.6
66.7
87.7
134.3
84.8
5
53
62
38
72
31
20
19
36
21
46
37
7
5
57
44
30
14
29
33
47
15
51
42
60
69
63
56
28
9
32
6
293.9
239.7
191.9
262.5
308.7
222.9
410.8
237.2
237.7
262.7
151.4
396.3
356.6
256.5
407.8
276.9
434.1
488.2
237.4
202.9
366.1
121.9
196.5
332.3
409.8
321.5
411.4
291.7
399.9
278.7
7
73.5
59.9
48.0
65.6
77.2
55.7
102.7
59.3
59.4
65.8
37.9
99.1
89.2
64.1
102.0
69.2
108.5
122.1
59.4
50.7
91.5
30.5
49.1
83.1
102.5
80.4
102.9
72.9
100.0
69.7
8
45
64
72
60
43
68
16
67
65
59
74
22
27
61
19
55
11
8
66
70
25
75
71
33
18
37
15
47
21
59
163
1
56
57
58
59
60
61
62
63
64
64
66
67
68
69
70
71
72
73
74
75
2
Barguna
Patuakhali
Bogra
Sherpur
Joypurhat
Dinajpur
Parbatipur
Thakurgaon
Ishurdi
Pabna
Serajgonj
Naogaon
Natore
Nawabgonj
Gaibandha
Kurigram
Lalmonir Hat
Nilphamary
Saidpur
Rangpur
3
795.1
1379.8
1960.7
1825.1
800.5
1244.2
662.0
609.2
429.4
1196.8
657.7
835.8
924.7
744.4
799.8
642.9
693.3
700.7
542.1
2010.7
4
72.3
125.4
178.2
165.9
72.8
113.1
60.2
55.4
39.0
108.7
59.8
76.0
84.0
67.7
76.7
58.4
63.0
63.7
49.1
182.9
5
52
12
4
6
48
16
64
70
75
17
65
45
34
55
49
67
61
59
73
3
6
291.8
377.7
492.7
279.2
270.0
402.1
290.2
357.8
269.6
392.2
328.6
317-4
325.9
252.6
308.5
423.1
205.4
279.0
316.2
462.3
7
73.0
94.4
123.2
69.8
67.5
100.5
72.6
89.5
67.4
98.1
82.2
79.4
81.5
63.2
77.1
105.8
51.4
69.8
79.1
115.6
8
46
24
7
52
56
20
50
26
57
23
34
39
35
63
44
12
69
53
40
9
164
Appendix 5: Ranking of Bangladesh Urban Centers (Hierarchy of Urban Centers)
SI. Name of the no. Urban Centers
Cumulative composite Indices Ail Bangladesh index (column 1,3,4, Municipalities 6,7,9,10-12,13,14,15) Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Dhaka (SMA )
Chittagong (SKA)
Khulna (SMA)
Rajshahi (SMA)
Cox's Bazar
Rangamati
B'Bria
Chandpur
Comilla
Feni
Lakshmipur
Chowmohani (Begamgong)
Noakhali (Sudharam)
Hobigong
Maulvi Bazar
Srimongal
Sunamgonj
Sylhet
Manikgonj
Munshigonj
Norsingdi
Faridpur
Rajbari
Gopalgonj
Madaripur
6,328.6
4,148.7
2,709.6
1,972.7
1,420.2
1,373.3
1,212.7
1,247.4
1,850.0
1,367.1
832.1
1,129,1
1,390.0
1,011.0
1,366.6
1,239.8
924.1
1,835.9
1,123.5
1,433.7
1,4301
1,537.2
1,015.1
1,139.7
896.4
421.9
276 6
180.6
131.5
94.7
91.6
80.9
83.2
123.3
91.1
55.5
75.3
1
2
3
9
23
25
37
33
11
26
73
44
92.7 24
67.4
91.1
82.7
61.6
122.4
74.9
95.6
95.3
102.5
67.6
75.9
59.8
58
27
35
68
12
45
21
22
19
57
43
71
Contd.
165
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
Jamalpur
Sherpur
Bajitpur
Bhairab
Kishorgonj
Gouripur
Mymonsingh
Muktagacha
Mohongonj
Netrokona
Gopalpur
Tangail
Barisal
Bhola
Jhalokati
Pirojpur
Jessore
Jhenaidah
Kotchandpur
MaHesHpur
Magura
Kalia
Narail
Bagarhat
Monglaport
(Rampal)
Satkhira
Chuadanga
Kumarkhali
Kushtia
Meherpur
1,076.3
930.9
1,097.0
832.2
1,242.1
1,316.2
1,568.9
1,155.6
1,328.6
1,088.8
1,065.6
2,049.0
2,253.4
979.1
1,249.6
1,221.4
1,783.9
1,446.0
1,168.7
1,005.8
1,628.6
920.5
1,055.5
1,025.8
1,019.6
988.5
1,144.7
1,256.9
1,877.4
1,211.5
71.8
62.1
73.1
55.5
82.8
87.7
104.6
77.1
88.6
72.6
71.1
136.6
150.2
65.3
83.3
81.4
118.9
96.4
77.9
67.1
108.6
61.4
70.4
68.4
68.0
65.9
76.3
83.8
125.2
80.8
50
67
47
74
34
29
18
40
28
48
52
8
6
64
32
36
13
20
39
59
16
69
59
55
56
61
42
30
10
38
Contd,
166
56
57
58
59
60
61
62
63
64
64
66
67
68
69
70
71
72
73
74
75
Barguna
Patuakhali
Bogra
Sherpur
Joypurhat
Dinajpur
Parbatipur
Thakurgaon
Ishurdi
Pabna
Serajgonj
Naogaon
Natore
Nawabgonj
Gaibandha
Kurigram
Lalmonir Hat
Nilphamaly
Saidpur
Rangpur
1,086.9
1.757.6
2,453.4
2,104.4
1,070.5
1,646.3
952.2
967.0
699.0
1,589.0
986.3
1,153.2
1,250.6
997.0
1,108.3
1,066.0
898.7
979.7
858.3
2,618.0
72.4
117.2
163.6
140.3
71.4
109.8
63.5
64.5
46.6
105.9
65.8
76.9
83.4
66.5
73.9
71.1
59.9
65.3
57.2
174.5
49
14
5
7
51
15
66
65
75
17
62
41
31
60
46
53
70
63
72
4
Bangladesh 1,500.0 100.0
Source: - BBS (1981) District census report, vol. 1-21, GOB.
- Bangladesh Municipalty Statistics (1977), National Institute
of Local government, Dhaka.
- BBS (1981) Report on Urban Areas, Government of Bangladesh.
167
BIBLIOGRAPHY
Abedin, J. (1969), Our Cities and Towns, NIPA, Dhaka, p. 64.
Abiodun, J.O. (1967), "Urban Hierarchy in a Developing Country",
"Economic Geography, Vol. 43, p. 353.
Ahmad, N. (1958), An Economic Geography of East Pakistan, Oxford
University Press, London, pp. 11-21.
Ahmad, N. (1957), "The Urban Pattern in East Pakistan", Oriental
Geographer, Vol. 1, p.28.
Ahmad, N. (1976), A New Economic Geography of Bangladesh, Vikas
Publishing House Pvt. Ltd., New Delhi, pp. 6-14.
Ahmad, Q. (1965), India Cities: Characteristics and Correlates,
University of Chicago, Department of Geography, Research
Paper no. 102, Chicago University of Chicago Press,
pp. 12-38.
Ahmad, Q.S. (1967), "Distribution Pattern of Urban Centers in
Pakistan'; Pakistan Geographical Review, Vol. 22, No. 1-8,
p. 21.
Alam, S.M. and Pokshishevsky, V.V. (1976), Urbanization in Develo
ping Countries, Osmania University, Hyderabad, p. 32.
Alamgir, M. (1973), Approaches towards research methodology on
problems of urbanization in Bangladesh, R.No. 15, October,
BIDS (Note: Immigrants from India was pronounced in early
1950's and the bulk of migratory currents took place in
1972 and afterwards, 0.7 million migrants from India,
were estimated in 1951), p. 9.
Alexander, J.W. (1954), "The Basic-Nonbasic concept of urban
Economic Functions", Economic Geography, Vol. 30, pp.
251-5.
Alexanderson, G. (1956), "City Forming and City Serving Production",
in H. Mayer and C.F. Kohn (eds.), Reading in Urban Geography,
Chicago, University of Chicago Press, pp. 163-9.
168
Ambrose, P. (1969), Analytical Human Geography, Longmants, London,
p. 10.
Anderson, Nels (1964), Our Industrial Urban Civilization, Asia
Publishing House, p. 3.
Andress, R.B. (1953), "Machanics of the Urban Economic Base:
Historical Development of the Base Concept", Land Economics,
Vol. 29, pp. 161-7.
Andrews, R.B. (1953), "Macharies of the Urban Economic Bases:
Historical development of the Basic concept". Land Economics,
Vol. 29, pp. 161-7.
Ayeni, B. (1979), Concepts and Techniques in Urban Analysis,
Choom helm, London, pp. 62-79, 92 and 165.
Bala, R. (1982), "New Towns in India: 1961, "Geoforum, Vol. 13,
N. 3, pp. 257-262.
Bala, R. (1986), Trends of Urbanisation in India, Prem Rawat,
Jawahat Nagar, Jaipur, p.7.
Batty, M. (1970), "An Activity Allocation Model for the Nottingha
mshire-Derbyshire subregion, "Regional studies. Vol. 4,
pp. 307-32.
Batty, M. (1970), "Dynamic Simulation of An Urban System", in
A.G. Wilson (ed.), London Papers in Regional Science,
Vol. 3, pp. 44-82.
Bartz, Albert, E. (1985), Basic statistical concepts, subject
Publications New Delhi, pp. 108-113.
Badcock, B.A. (1970), "Central place evolution and Net Development
in South Aukland 1840-1968: A system Analytic Approach",
NewZealand Geographer, Vol. 26, p. 109.
Bannister, G. (1977), "Space-time Components of Urban Population
Change", Economic Geography, Boston University, p. 39.
B.B.S. (1984), National Census-District Volumes, Government
of Bangladesh, p. 9.
169
B.B.S. (1987), Bangladesh Population Censes-1981, Government
of Bangladesh, Ministry of planning, Dhaka, Vol. 1-21
various pages.
Beattie, E.F., and Watts, H.D. (1983), "Some Relatioinship Between
Manufacturing Activity and Urban System: an Exploratory
Study, "Geoforum, Vol. 14, No. 2, p. 38.
Beckmann, M.J. and McPherson, J.C. (1970), "City Size Distribution
in a central Place Hierarchy: An Alternative Approach",
Journal of Regional Science, Vol. 10, No. 1, p. 11.
Berg, L.V.D. (1987), Urban System in a Dynamic Society, Gower
Publishing Company Limited, Hampshire, England, p.2.
Bergel, E.E. (1955), Urban Sociology, McGraw Hill Book Company
Inc. New York, pp. 8-10.
Bernet, A.J. (1937), Cultural Relations between India and Java,
Kempors, pp. 102-12.
Berry, B.J.L. (1961), "City Size Distributions and Economic
Development", Economic Development Cultural Change,
Vol. IX, pp. 38 and 575-88.
Berry, B.J.L. (1964), "Cities as Systems within Systems of Cities",
Papers and Proceeding of the Regional Science Association,
Vol. 13, pp. 21-34.
Berry, B.J.L. (1971a), "City Size and Economic Development",
in L. Jakobson and V. Prakash, (eds.), Urbanizations
and National Development, Saga publications, Beverly
Hills, Calif., pp. 18 and 111-55.
Berory, B.J.L. (1971b), "The Geography of the United States in
the 2000", Ekistics, pp. 339-51.
Berry, B.J.L. (1971c), "Contemporary Urbanization Processes",
in U.S. National Academy of Science, Geographical Pers
pectives on Urban Problems, pp. 94-107.
170
Berry, B.J.L. (1972a), "Latin Structure of the American Urban
System", in B.J.L. Berry, eds., City Classification
Handbook, Wiley Interscience, New York.
BERRY, B.J.L. (1972b), "Hierarchical Diffusion: The Basis of
Developmental Filtering and Spread in a System of Growth
Centers", in N.M. Hansen, ed., Growth Centers in Regional
Economic Development, The Free Press, New York, pp.
108-38.
Berry, B.J.L. (1973). "Growth Centers in the American Urban
System", Vol. 1, Community Development and Regional
Growth in the 60's and 70's, Ballinger, Cambridge, Mass,
p. 311.
Berry and Garrison, (19 58), "Recent Developments of Central
Place Theory", Papers and Proceedings of Regional Science
Association, Vol. 4, pp. 107-20.
Berry, B.J.L., and Horton, F.E., (1970), Geographic prespectives
on Urban Systems: with integrated Readings: Prentice-
Hall, New Jersey, pp. 20-35, 69 and 468.
Berry and Garrison (1958), "Alternative Explanations of Urban
Rank-Size relationship", AAAG, Vol. XLVIII (March),
pp. 83-91, and 304-11.
Best, J.W. and Khan,. J-V. (1986), Research in Education, Prentice-
Hall of India, New Delhi, p. 223.
Beumer, L., et. al. (1978), "A Study of the Formal Structure
of J.W. Forrester's Urban Dynamics Model", Urban Studies,
Vol. 15, p. 109.
Bogue, D.J. (1959), The Population of the United States, The
Free Press, Glencoe, p. 4 87.
Bourne, L.S. (1975), Urban System: Strategies of Regulation-
A Companison of Policies in Britain, Sweden, Australia,
and Canada, Clarendon Press, Oxford, p. 11.
171
Bourne, L.S. (1982), Internal Structure of the city: Reading
on Urban form, growth, and Policy, Oxford University
Press, New York, pp. 104-48.
Bourne, L.S. and Murdic, R.A. (1972), "Interrelationship of
Social and Physical space in the city; A Multivariate
Analysis of Metropolitan Toronto", The Canadian Geographer,
Vol. 26, No. 3, pp. 211-29.
Bourne, L.S. and Simmons, J.W. (1978), Systems of Cities: Reading
on Structure, growth, and Policy, Oxford University
Press, New York, pp. 1-25.
Borchert, J.H. (1986), "Urban Systems in transition and govern
ment actions", Netherland Geographic Studies, Vol. 16,
pp. 222-230.
Boventer, E. Ven (1978), "Bandwagon Effects and Product Cycles
in Urban Dynamics", Urban Studies, Vol. 15, pp. 261-272.
Breese, G. (1966). Urbanization in Newly Developing Countries,
Bureau of Urban Research, Princeton University, Prentice-
Hall, Inc. Englewood Cliffs, N-J., London, p. 6.
Browning and Gibbs (1966), "Some Measures of Demographic and
spatial Relationship Among Cities" in Gibbs, Urban
Research Method, D. Van Nortrand Company, New York,
pp. 436-57.
Bhat, L.S. et. al., (1976), Micro-level Planning: A Case Study
of Karnal Area, Haryana-India, K.B. Publications, New
Delhi, pp. 60-61.
Carrothers, G.A.P. (1956). An Historical Review of the Gravity
and Potential Concept of Human interaction", Journal,
American Institute of Planners, Vol. 22, p. 95.
Center for Urban Studies (1978), Study of an inner-city slums
in Dhaka, United Nations Center for Housing, Building
and Planning, Dhaka, pp. 17-18.
172
Chamber's Twentieth Century Dictionary (1965), Bombay, p. 31.
Chatterjee, S.P. (1962), "Regional Pattern of Desnity and Distri
bution in India", Geographical Review of India, Vol.24,
pp. 1-28.
Childe, V.G. (1936), Man Makes Himself, London, p. 69.
Clark, P.J. and Evans, F.C. (1954), "Distance to Nearest Neighbour
as a Measure of Spatial Relations in Population, "Ecology,
Vol. 35, October, pp. 445-453.
Chandra, R.C. (1986), A Geography of Population: Concepts, deter
minants and Patterns, Kalyani Publishers, New Delhi,
pp. 65-66.
Chapman, G. (1983), "The Growth of Large Cities in India 1961-
1971, Geoforum, Vol. 13, No.2, p. 64.
Chapman, G. and Wanmali, S. (1981), "Urban-Rural Relationships
in India: a Macro-Scale Approach Using Population
Potentials", Geoform, Vol. 12, No. 1, pp. 19-43.
Chen, K. (1972), Urban Dynamics: Extensions and Reflections,
(ed). Sand Francisco Press, San Francisco, pp. 213-16.
Chisholm, M. (1972), "Macro-and Micro-Approaches to Urban Systems
Research," The Geographical Journal, Vol. 138, No. I,
pp. 60-63.
Chisholm, M. and Manners, G. (1971) "Geographic Space: A New
Dimension of Public Concern and Policy", in M. Chishelm
and G. Manners, (eds.), Spatial Policy Problems of the
British Economy, Cambridge University Press, Cambridge,
pp. 1-23.
Chowdhury, R.H. and Curlin, G. (1975), "Dynamics of Migration,
Development studies. Vol. Ill, No. 2 (April), pp. 181-
230.
Chowdhury, Osman, H. (1983), "Agricultural underemployment,
industrial Growth, Real wage and Labour Absorption in
Bangladesh Economy: 1952-1978", A paper presented in
a BIDS in-house seminar, May, Dhaka, p. 14.
173
Coedes, G. (1968), The indiaized States of South-east Asia,
East-West Centers Press, 3rd ed. Honolulu, pp. 42-
9.
Crossby, T. (1986), Architecture: City Sence (London Studio
Vista Limited, Blue Star House, High Gate Hill London
N19), p. 19.
Cressey, Paul F. (1930), The Succession of Cultural Groups
in the City of Chicago, Unpublished Ph.D. dissertation.
University of Chicago, p. 227.
Curry, L. and Mackinnon, R.D. (1974) "Aggregative Dynamic urban
Models Oriented Towards Policy", "Report to the Ministry
of State for urban Affairs by the Center for Urban
and Community Studies, University of Toronto, p. 351.
Dacey (1962), "Analysis of Central Places and Point Patterns
by a Nearest Neighbour Method", in K. Norberg (ed.)
Proceedings of the I.G.U. Symposium in Urban Geography,
pp. 55-75 and 122-94.
Davis,K. (1961), "Forward: Urban Research and Its significance",
in Gibbs, J.A., Urban Research Method, New York, pp.
V-XVII and 427.
Davis, K. and Golden, H.H. (1965), "Urbanization and the Develo
pment of pre-industrial Area", in the Hall and Reiss
(eds.) Cities and Society, New York, pp. 41-53.
Day, L.H. (1979), "What will be a ZPG Society be Like?" Popula
tion Bulletin, Vol. 33, No. 3, pp. 1-38.
Dickinson, R-E. (1967), "The Scope and Status of Urban Geography:
an assessment", in Mayer, H.M., and Kohn. C.F. (eds.),
Reading in Urban Geography, Allahabad, pp. 10-11.
Dickinson, Robert E. (1956). City Region and Regionalism, London:
Routledge and Kagan Raed, p. 17.
174
Domanski, R. (1973), "Structure, Law of Motion and Optimal
Path of Growth of Complex Urban System", Economic
Geography, Vol. 49, pp. 37-46.
Domros, M. (1984), "The Population Explosion and Demographic
Change in India", Applied Geography and Development,
Vol. 24, pp. 77-90.
Duncan, O.D. and Shindman (1951), "Optimum Size of Cities",
(eds.), J.J- Spengler and O.D. Duncan; Demographic
Analysis, Selected Readings, Free Press, pp. 85-87.
Elahi, K.M. (1985), Urbanization in Bangladesh: A Historical
Background, 4th NGC, Bangladesh National Geographical
Association, Dhaka, pp. 18-19.
Fischer, C.S. (1976), The Urban Experience, New York, Harcourt
Brace Javanovich, p. 3.
Forrester, J.N. (1969), Urban Dynamics, M.I.T. Press, Chambridge,
Mass, p.9.
Friedman, J. (1966), "Two Concepts of Urbanization: A Comment",
Urban Affairs, Quarterly, Vol. 1. p. 17.
Friedman, J. (1972), "The Spatial Organization of Power and
the Development of Urban System", University of
California, School of Architecture and Urban Planning.
Los Angeles (Mimeo), p. 108.
Friedmann, J. and Wilson, R.H. (1977), A Critical Look at Urban
Dynamics, The Urban Institute, Washington, p. 28.
Gallion, A.B., and Eisner, S., (1963), The Urban Pattern-City
Planning and design, New York, p. 3.
Ganguly, D.C. (1943), "Bengalis Outside Bengal", History of
Bengal, Vol. 1, Chapter XVII, University of Dhaka,
pp. 161-8.
Garner, B.J. (1970), "Aspects and Trend of Urban Geography"
in Putnam, R.G., Taylor, F.J. and Kattle, P.G. (eds),
Geography of Urban Places, Mathuen, pp. 4-10.
175
Gamier, B-J., and Chabot, G., (1967), Urban Geography (eds),
London, p. 113.
Gibbs, J.P. (1961), Urban Research Methods (eds), VanNostrand
Company, Inc. Princeton, New Jersey, New York, pp.
14 and 27.
Gist, N.P., and Halbert, L.A., (1954), Urban Society, Thomas
Y. Crowell Company, New York, pp. 77-94 and 149-175.
GOB (1983), Bangladesh; Ministry of Information and Broad Casting,
Dhaka, p. 4.
GOB (1981), Pourashava Manual, Dhaka Municipalty, Dhaka, p.
50.
Gosal, G.S. (1972), "Internal Migration in India - A Regional
Analysis", Indian Geographical Journal, Vol. 36, pp.
173-203.
Gosal. G.S. and Krishan, G. (1975), "Patterns of Internal Migra
tion in India", in L.A. Kosinski and R.M. Prothero
(eds.). People on the Move, Methuen and Co. Ltd., London,
p. 200.
Grove, D., and Huszar, L. (1964), The Towns of Ghana: The Role
of Seirvice Centers in Regional Planning, Accra, p. 63.
Hartshorn, T.A. (1980), Interpreting the City: An Urban Geography,
John Wiley and Sons, New York, pp. 2, 15, 90, 206-8,
and 286.
Hauser Philip M., and Duncan, Otis D. (19 59), The Study of Popula
tion An Inventory and Appraisal (eds.). The University
of Chicago Press, p. 48.
Harris, Chauncy D. and Edward L. Ullman (1945), "The Nature
of Cities". Annals of the American Academy of Political
and Social Science, Vol. 242 (November), pp. 114-35.
Hiorns, F.R. (1956), Town Building in History, New York, p 13
176
Hoch, I. (1972) "Income and City Size", Urban Studies, Vol.9,
pp. 299-328.
Hoselitz, B.F. (1954-55), "Generative and Paracitic Cities",
in Economic Development and Cultural Change, Vol.3,
pp. 278-294.
Hoyt, H. (1961), "The Utility of the Economic Base Method in
Calculating Urban growth:, Land Economics, Vol. 37, pp. 51-8.
Hudson, J.C. and Fowler, P.M. (1969), The Concept of Pattern
in Geography, University of Iowa, Department of Geography,
Discussion Paper Series, pp. 1-11.
Hughes, J.W. (1972), Urban Indicators, Metropolitan Evolution
and Public Policy, Center for Urban Policy Research,
Rutgers University, New Brunswick, N.J., p. 139.
Hunter, W.W. (1964), A Statistical Account of Bengal, Vol. XII
( rtaldah, Rangpur, Dinajpur), London, p. 63.
Islam, N. (1976), The Urban Poor in Bangladesh, CUS, p. 18.
Isard, W. (1956), Location and Space Economy, New York, John
Wiley, p. 196.
Jefferson, Mark (1931), "Distribution of the World's City Folks",
Geographical Review, Vol. XXI, (July), pp. 453 and
537-66.
Johston, R.J. (1977), "Reading Urban Origins, Urbanization and
Urban Pattern", Geography, Vol. 66, p. 28.
Jones, H.R. (1981), A Population Geography, Harper and Row
Publishers, London, p. 201.
Juppenlatz, M. (1984), "An information system for Urban modelling",
ITC Journal, Vol. No. 2. pp. 89-94.
Kenneth, S. (1980), The Changing Distribution of Sex Ratios
in the British Urban System", Geoforum, Vol.11, pp.
31-41.
177
Keyfitz, Nathan (1972), "Population Wares", Greville, T.N.E.
(eds.) Population Dynamics, Proceeding of a Symposium
conducted by the Mathematics Research Center, The
University of Wisconsin, Madison, June 19-21, Academic
Press, New York, London, pp. 1-3.
Keyfitz, Nathan (1980), "Do cities Grow by Natural Increase
or by Migration?", Geographical Analysis, Vol. 12,
No. 2, p. 14 2.
Khan, F.K. (1959), An Introduction to Economic Geography, New
Age Publications, Dhaka, pp. 407-8.
Kibel, B.M. (1972), "Simulation of Urban Environment", Commission
on College Geography, Technical Paper, No. 5, AAG,
pp. 21-35.
King, L.J. (1962), "A Quantative expression of the Pattern of
urban settlements in selected Areas of United States",
TESG, January, pp. 1-7.
King, L.J. (1969), Statistical Analysis in Geography, Prentice,
Hall, Inc. Englewood Cliffs, pp. 22-48.
Krishan, G. and Bala, Raj (1982), "Urbanization in Border Regions;
A case study of India's Border Districts Adjoining
Pakistan", The Geographical Journal, Vol. 148, pp.
43-49.
Lee, E.S. (1969), "A Theory of Migration", Demography, Vol.3,
p. 49.
Logan, M.I. (1966), The Location Behaviour of Manufacturing
Firms in urban Area, AAAG, Vol. 56, pp. 451-66.
Losch, A. (1954), The Economics of location, translated from
the second revised eddition by william, H. Waglom with
the assistance of Wolfgang, F, Stolper, New Haven:
Yale University Press, pp. 59-72.
Lowry, I.S. (1964), A Model of Metropolis, Santa Monica,
California, Rand Corporation, pp. 311-29.
178
Lucas, D. (1979), Beginning Population Studies, Australian
National University Press, Canberra, pp. 40-48.
Mabogunje, A.L. (1973), Towards an Urban Policy in Nigeria,
An address presented at the 17th Annual Conference
of the Nigerian Geographical Association, Lugos, December,
p. 4.
Mahmood. A. (1977), Statistical Methods in Geographical Studies,
Rajesh Publications, New Delhi, pp. 145-60.
Marshall, J.V. (1971), The Location of Service Town: An Approach
to the Analysis of Central Systems, University of Toronto
Press, Toronto, p. 49.
Majumder, R-C. (1944), Hindu Colonies in the Far East, General
Printers, Calcutta, p. 19.
Mandal, R.B. (1982), Statistics for Geographers and Social,
Scientists, Concept Publishing Company, New Delhi,
pp. 497-99.
Mayer, H.M. (1970), "Cities in Urban Geography", in patnam,
R.G., Taylor, F.J. and Kettle, P.G. (eds.), A geography
of Urban Places, Toronto, p. 11-23.
Mehta, B.C. (1978), Regional Population Growth: A Case Study
of Rajasthan, Research Books, C-150, Sunder Marg, Tilak
Nagar, Jaipur, pp. 122-125.
Misra, H.N. (1984), "Urban System of a Developing Economy",
International Institute for Development Station,
Allahabad, pp. 1-251.
Morrill, Richard L. et.al. (1984), "The Settlement System of
the United States", Urbanization and Settlement System"
International Perspectives, Oxford University Press,
New York, p. 37.
Morris, P.A. (1972), Population Movement and the Shape of Urban
Growth; Implications for Public Policy, Rand Corporation,
Santa Monica, Calif, pp. 29-33.
179
Morrison, W.I. (1973), "The Development of an Urban Inter-industry
Model 1, Building the Input-Output Accounts", Environment
and Planning, Vol.5, pp. 569-83.
Mumford, Lewis (1961), The City in History, London, pp. 13-64.
Noble, A.G. and Dutt, A.K. (1978), Indian Urbanization and Plann
ing; Vehicles of Modernization, Tata Machraw Hill Publi
shing Company Limited, New Delhi, pp. 1-10.
Ogunsanya, A.A. (1985), "Generating Urban Traffic Flow Data
for Urban transport Studies in Developing Countries:
An Approach," Geoforum, Vol. 16, No. 4, pp. 403-412.
Orcutt, G.H., et al. (1961), Micro-analysis of Socio-economic
system, A Simultation Study, New York, pp. 137-49.
Paelinck, J. (1970), "Dynamic Urban Growth Models", Papers
Regional Science Association, Vol. 24, pp. 25-37.
Paelinck, J.H.P. (1974), "Alternative Methods for the Study
of Urban Dynamics", in Funck, R. (ed.) Recent Developments
in Regional Science, U.S.A., pp. 33-49.
Pahl, R. (1971), "Povery and the Urban System", in M. Chisholm
and G. Manners, (eds.), Spatial Policy Problems in
the British Economy, pp. 126-4 5.
Parr, J.B. (1970), "Models of city size in an Urban System",
Papers of the Regional Science Association Vol, 25,
pp. 221-53.
Pandey, P. (1973), "Urban Hierarchy in Chotanagpur", in Singh,
R.L. (ed.). Urban Geography: In Developing Countries,
BHU, Varanasi, India, pp. 306-315.
Pedersen, P.O. (1970), "Innovation Diffusion within and between
National Urban Systems", Geographical Analysis, Vol.2,
p. 223.
Peterson, Willian (1961), Population, The MaCmillan Company,
New York, pp. 179-201.
180
Pred, A.R. (1966), The Spatial Dynamics of U.S. Urban Industrial
Growth 1880-1914, MIT, Press, Cambridge, Mass, pp.
102-21.
Pred, A. (1967), Behavior and Location Part-I, Lund Series in
Geography, Series B, No. 27; Land, Gleerup, pp. 61-
83.
Preston, S. (1975), Mortality Patterns in National Populations,
Academic Press, New York, pp. 231-248.
Pumain, D. et al. (1984), "Towards a modelling of intra-urban
dynamics", Espace Geographigue, Vol. 13, No. 2., pp.
125-135 and 308-325.
Queen, S. and Carpenter, D.B. (1953), The American City, New
York, p. 29.
Rasheed, S. (1972), Bangladesh: Bhogolic Parichoai, Center for
Urban Studies (C 113), D.U., Dhaka, pp. 6-7.
Rashid, H. (1977), Geography of Bangadesh, University Press
Ltd. Dhaka, pp. 15, 165-73 and 292.
Rashid, H. (1981), An Economic Geography of Bangladesh, University
Press Ltd., Dhaka, Bangladesh, pp. 14-21.
Ravenstein, (1885), "The Laws of migration". Journal of the
Royel Statistical Society, Vol. 48, pp. 167-235.
Rejuan, H. (1982), "Morbidity and Mortality of a semi-urban
Community", Unpublished Master's thesis in Demography,
Dhaka University, p. 5.
Richardson, H.W. (1973a), "Theory of the Distribution of City
Sizes: Review and Prospects", Regional Studies, Vol.7,
pp. 239-51.
Richardson, H.W. (1973b), The Econom'c of Urban Size, D.C. Health,
London, pp. 128-38.
Roberts, Bryon (1982), "Cities in Developing Societies", Introduc
tion to the Society of Developing Societies, in Alvi
and Teodor Shamim, McMillion Press Ltd, London, pp.3&371.
181
Roteinis, v., and Calef, W. (1955), Notes on the Basic-Nonbasic
Employment Ratio, Economic Geography, Vol. 31, pp.
17-20.
Shafer, T.W. (1977), "The Dynamics of Economic growths," Urban
Growth and Economics, Reston Publication Company, Inc.
A pentice Hall Company, Reston, Virginia, pp. 61-78.
Shafi, M. (1984), Agricultural Productivity and Regional Imba
lances: A study of Uttar Pradesh, Concept Publishing
Co. New Delhi, pp. 69-131.
Sharma, A.N. (1981), Spatial Approach for District Planning:
A Case Study of Karnal District, New Delhi, ; pp. 1
and 78.
Sharlin, A. (1978), "Natural Decrease in Early Modern Cities:
A Reconsideration", Past and Present, Vol. 79, pp. 126-38.
Simmons, James. W. (1984), "The Canadian Urban System", Urbaniza
tion and Settlement Systems, Oxford University Press,
New York, pp. 10-10 and 31-38.
Singh, O. (1975), "The Size and Spacing of towns in the umland
of Allahabad", The Geographer, Vol. XXII, pp. 4 5-55.
Singh, A.M. (1976), Neighbourhood and Social network in Urban
India, Marwah, New Delhi, pp. 193-94.
Singh, J.P. (1979), "Problem of Urbanization in India", Mainstreeim,
Vol. XVII, No. 25, Feb. 17, Delhi, pp. 53-54.
Sinha, B.N. (1970), SIRSI: An Urban Study-in application of
Research, model, Karnataka University, Dharwar, p. 66.
Smailes, A.E. (1981) The Geography of Towns, London, pp. 11-23.
Smith, M.W. (1955), Technological Change and Social Disorgani
zation: Bibliography on Asia and Pacific, Paris, Conscil
International des Sciences Socialas, Bureau International
de recterche Sur les Implications Socialas der Projrer
Technique, p. 419.
182
Smith, W. (1955), "The Location of Industry", Transactions
Institute of British Geographers, Vol. 21, pp. 1-18.
Stamp, Sir Dudly and Audrey N. Clark (1979), A Glossary of Geo
graphical Terms, Longman, London, p. 4 99.
Stewart, J.Q. (1948), "Demographic Gravitation: Evidence and
Application", Sociometry, Vol. 11. pp. 31-58.
Stewart (1958), "The Size and Spacing of Cities", Geographical
Review, Vol. XLVIII (April), pp. 222-45.
Symanski, R. and Newman, J.L. (1973), "Formal, Functional, and
Nodal Regims; Three Fallacies", The Professional Geo
grapher, Vol. 25, No.4, pp. 131-8.
Thakur, B. (1980), Urban Settlements in Eastern India-Entropy
Changes and Pattern Analysis, Concept Publishing Caompany,
New Delhi, pp. 1-24, and 44-56.
Thompson, W.R. (1972), "The National System of Cities as an
object of Public Policy", Urban Studies, Vol. 9, pp.
99-116.
Thomas, E.N. (1961), "Toward an expended central Place Model",
Geographical Review, Vol. 51, pp. 400-11.
Thomas, E.N. (1965)." Stability of Distance-Population Size
Relationship", Proceeding of the I.G.U Symposium in
Urban Geography, (ed.) K. Norberg; King, L.J., The
Spacing of Urban Places in the United States, Unpubli
shed Ph.D.Thesis, Department of Geography, State Univer
sity of Iowa, 1960; Vaicich, G., "An Analysis of Spacing
of the Small Towns in Iowa", Unpublished Ph.D. Thesis,
Department of Geography, State University of Iowa,
1960, pp. 13-29.
Tkachenko, L. Ya (1984), "Physical-geographic aspects of the
study of urban area", Soviet Geography-Review and Trans
lation, Vol. 25, No. 5, pp. 248-252.
183
Trewartha, G.T. (1969), A Geography of Population; World Pattern,
John Wiley and Sons Inc., New York, p. 137.
Ulman, E. (1956), "The Role of Transportation and the Bases
for Interaction", in Man's Role in Changing the Face
of the Earth, University of Chicago, pp. 862-80.
Vining, R. (1955), "A Description of certain Spatial Aspects
of an Economic system", Economic Development and Cultural
Change, Vol. 3, pp. 147-96.
Von Borenter, E. (1973), "City-Size Systems: Theoretical Issues,
Empirical Regularities and Planning Guides", Urban
Studies, Vol. 10, pp. 145-62.
Wadia, Krishnan and Mukerjee (1935), Record in Geological Survey:
India. Vol. LXIII, Part 4, Calcutta, pp. 363-91.
White, Harvey (1963), College Physics, Van Nestrand, New York,
in Hartshorn (1980), Interpretating the City, p. 97.
Whyte, A. (1985), "Ecological approaches to urban systems: a
retrospective and propsective look". Nature and Resources,
Vol. 21, No. 1, pp. 13-20.
Wrigley, E.A. (1969), Population and History, New York, p. 18.
Yixing, Zhou and Z. Qin (1984), "On the determination of City
Character in Urban Planning, "Scientia Geographic Since,
Vol. 4, No. 1, pp. 29-37.
Zelinsky, W. (1971),"The Hypothesis of the Mobility Transition",
Geographical Review, Vol. 61, p. 223.
Zipf, G.K. (1949), Human Behaviour and Principle of least Effort,
New York: Addison Wesley, Cambridge, p. 189.
