Cambridge Books Online
http://ebooks.cambridge.org/
World Geomorphology
E. M. Bridges
Book DOI: http://dx.doi.org/10.1017/CBO9781139170154
Online ISBN: 9781139170154
Hardback ISBN: 9780521383431
Paperback ISBN: 9780521289658
Chapter
2 - Continental drift and plate tectonics pp. 12-29
Chapter DOI: http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge University Press
Continental drift and plate
tectonics
Science fiction is a literary form which enables authors
to imagine amazing and sometimes frightening ideas of
what life may be like on other planets or even on earth at
some time in the past or future. Sometimes fact is
stranger than fiction for one of the most intriguing
stories of scientific discovery concerns our own planet,
Earth, and the way the continental areas have evolved
and assumed their present distribution.
For a long period geographers and geologists thought
that there was a strong degree of permanence in the
distribution and form of the continents and ocean
basins. Although the highest mountain on land and the
deepest parts of the sea were of great interest, it was
significant that the average level of land occurs at 870 m
above present sea level and that the average depth of the
sea is 3800 m below sea level. This average picture,
shown on the hypsometric diagram (Figure 2.1), also
gives a good idea ofthe general shape of the ocean basins
and their resemblance to a soup dish. The ocean waters
slightly overfill the dish and extend on to the gentle
slope of the rim. This gentle slope, known as the
continental shelf, is bounded on the ocean side by a
relatively sharp descent to the ocean depths of the
abyssal plain. The North Sea and the shallow seas
around Great Britain are a good example of the continental
shelf at the present time. However, epicontinental
seas such as this have varied considerably throughout
geological time and many of the familiar
sedimentary strata were laid down in them.
Besides the clear difference in relief, it has been
appreciated for the last 50 years that the major landmasses
were predominantly composed of lighter mater-
12
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Continental drift and plate tectonics 13
8840 m
Sea Level
Figure 2.1. The extent of global relief lying at different elevations is
shown by the hypsographic curve. The mean continental level is
370 m and the mean oceanic level is — 3730 m.
i 35%
Copper
65%
Mercury
35%
65%
Density 13.6gm/cc
Ocean
Basin
Denser rocks
Deeper roots
Figure 2.2. The principle of isostasy states that the Earth's crust is
in a state of buoyant equilibrium. Blocks of the same density but
different lengths will float at different levels; in the case of the
Earth's crust the lighter sial 'floats' above the denser sima, but high
mountains must be compensated for by deep roots.
ials such as granites, rich in silica and aluminium.
Continental material, therefore, received the acronym
'sial', and as it had an average specific gravity of only 2.7
it was thought to 'float' on the denser material which
made up the seafloorand which underlies the continental
material at depth. The ocean floor material, rich in
silica and magnesium became known as 'sima' and had a
specific gravity of between 2.9 and 3.0. The discontinuity
between granitic and basaltic material has already
been mentioned in the previous chapter as lying
between 20 and 25 km beneath the surface and is also
known as the Conrad discontinuity.
In order to meet the physical requirements imposed
by gravity, and for high mountains and deep sea basins
to exist, the concept of 'isostasy' was introduced by the
American geologist Dutton in 1889. If blocks ofwood of
different length arefloatedupright in water, or if blocks
of copper are similarly floated in mercury, they assume
different elevations in proportion to their length. They
also protrude by different amounts into the supporting
liquid (Figure 2.2). If this concept is applied to sial and
sima, it becomes immediately apparent that the higher
the mountain mass, the deeper roots it must have
penetrating down into the earth.
Evidence for deep roots beneath large mountain
masses was first noted by the French in a scientific
expedition to the Andes in thefirsthalf ofthe eighteenth
century. A smaller than expected deflection of the
plumb-line was observed in the Indo-Gangetic plain in
response to the considerable mountain mass of the
Himalayas. Once adjustments have been made for
elevation and for the effects of the pull of the Sun and
Moon, the values of gravity should approximate to
calculated values for the idealised ellipsoid Earth. That
they do not is further indication that material of lower
specific gravity penetrates deeply below major mountain
ranges. Evidence from earthquakes suggests that
the roots of mountains extend to depths of 40 km or
more.
Before the ideas of continental drift were put forward
by Wegener in his paper of 1912 and his book of 1915
The origin of continents and oceans, geologists were in a
great dilemma. It was apparent that there had been
considerable crustal shortening as could be seen where
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
14 Continental drift and plate tectonics
the rocks were folded and overthrust. This led them to
the conclusion that the Earth had contracted and that
the mountains represented the wrinkled crust of the
shrinking sphere. A further difficulty was that the
climatic conditions in which rock strata had been laid
down had varied greatly with the passage of time; either
the climate had changed greatly or the continent might
have had a different disposition from that of the present
day, but how could a continent move?
One ingenious, but unsupportable, scheme was that
of Lothian Green. Shrinkage of the globe, he suggested,
would lead to a tetrahedral shape with the oceans
occupying the faces of the tetrahedron and the continents
the edges. Although imaginative, this idea cannot
stand up to several arguments; it would be gravitationally
unstable and isostatic readjustments would take
place causing the corners to sink back into the Earth,
and, as can be seen from space, the major relief features
of the Earth at true scale scarcely diverge from the
circumference of the globe.
The absence of continental material, sial, from large
areas of the Earth's surface gave rise to some other
interesting theories. One of the most intriguing was that
proposed by George Darwin, son of the famous natural
scientist Charles Darwin. He suggested that the sialic
rocks, missing from the whole of the Pacific basin, had
been removed from the Earth and now formed the
moon. The Pacific basin was the scar left behind when
the Moon was separated from the Earth. Darwin
calculated that the combined effects of the rotation of
the Earth and the Sun's attraction would produce a tidal
resonance which would increase until a mass separated
from the Earth. The attraction of this idea was that the
Moon has been found to be moving further away from
the Earth with the passage of time. However, the tidal
resonance theory of the origin of the moon and the
resultant lack of sial on the areas of the Pacific Ocean
floor was disproved because internal friction would
prevent the tidal resonance proposed by Darwin.
Alternative theories now suggest that the Moon is
most probably a captured body and that this capture
must have taken place simultaneously with the formation
of the Earth. Darwin's idea that the removal of
material from the Pacific resulted in the movement of
the other continental masses to give the ring of mountains
and volcanoes which surround the Pacific must
also be abandoned, and some other explanation sought
for the so-called ring of fire.
Evidence for continental movement has been
accumulating since the beginning of the seventeenth
century when Sir Francis Bacon noted the complementary
nature of the coasts of Africa and South America. A
Frenchman in the seventeenth century concluded that
America was separated from the Old World at the time
of the biblical Flood; other scientists denied the whole
affair and suggested that the Atlantic resulted from the
foundering of the mythological Continent of Atlantis.
Many geological events were seen to have taken place in
a catastrophic manner in the early period of geological
studies. The correspondence of geological features on
both sides of the Atlantic, including the distribution of
similar fossil plants in the coal measures of Europe and
North America, was first described in detail in 1858.
Continental drift
By the first decade of the twentieth century sufficient
information had been accumulated for more definite
theories on the movement of continents to be put
forward. Two Americans were first in the field: F.B.
Taylor (1908) and J.B. Baker (1911) used the idea of
continental drift to explain the origin of mountains and
the correspondence of relief on either side of the
Atlantic. However, the most famous protagonist during
this period was undoubtedly Alfred Wegener, who was
not a geologist but a meteorologist. Wegener had
adopted the idea of continental drift to account for the
different climates which had affected continents in
earlier geological times. He had amassed a wealth of
biological, palaeontological, palaeoclimatic, tectonic
and geophysical evidence for the existence of continental
drift. Unfortunately, the concept was too much for
his contemporaries, who picked holes in his arguments
and rejected the hypothesis. In spite of some excellent
positive evidence, his proposals really foundered on
disbelief, as nobody at that time could envisage a force
or mechanism which could move a continent across the
face of the Earth.
The evidence collected by Wegener suggested to him
that the present distribution of the continents is derived
from the disruption ofa super continent which he called
'Pangea' (all lands) which had existed in the late
Carboniferous (Figure 2.3). The break-up of this large
landmass took place in the Mesozoic, with the core areas
of the future continents moving towards their present
positions. Although Wegener and some of the early
workers were trying to show how different climate
might have affected the five continents, their maps did
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Continental drift 15
PANGEA
Figure 2.3. A best-fit reconstruction of the continents to form
'Pangea' (all-lands) as it may have been in Permian times. Laurasia
and Gondwana were separated subsequently by rifting and
movement of the tectonic plates.
not show Europe and North America sufficiently near
the equator to have tropical forests. Wegener was well
aware of the northward movement of Africa and India
as he invoked a force called 'polflucht' to account for
their motion. Such a force does in fact exist, but it would
have to be many million times more powerful before it
could move a continent. Additionally, such a force
would not be adequate to account for the generally
westward movement of North and South America, and
certainly could not explain any reversals of movement
which are thought to have occurred.
Other proponents of the continental drift theory,
such as du Toit, distinguish between a northern group
of continental nuclei, which they called 'Laurasia' and a
southern group called 'Gondwana' (Figure 2.4). The
name Laurasia is formed from elements of words
describing part of North America and Asia. The name
Gondwana is taken from a place in India, Gondwana,
where fossil plants of the Glossopterisflorawere first
described, only to be found later in South America,
Africa, Australia and Antartica. Between Laurasia and
Gondwana a geosynclinal area developed beneath a sea
known as Tethys. Into this subsiding zone the deposits
were laid which eventually became the fold mountains
of the Alps and the Himalayas.
Although opinion generally swung away from the
idea of continental drift as a practical explanation of
continental distribution, the idea was kept alive by one
or two active proponents. In South Africa, Alexander
du Toit remained convinced of the possibility of continental
drift and wrote articles and a book supporting the
theory. In Britain, Anthony Holmes can be credited
with being the first to suggest a possible mechanism for
moving continents. After working on methods of dating
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
16 Continental drift and plate tectonics
Pre-drift reconstruction blocks of
age > 1 700 million years (dotted areas)
Figure 2.4. Continental nucleii - cratons of 1700 million years ago
reassembled into their positions before disruption of Pangea. (After
Hurley and Rand, 1969, Science 164, 1229.)
rocks by measuring radioactive decay, he proposed that
the heat liberated by radioactive disintegration of
mineral deep in the Earth's crust would cause convection
currents which might raft the continents along.
In spite of the convincing nature of the evidence
collected by those scientists who favoured continental
drift, the geological establishment of the time was very
reluctant to accept the concept of moving continents.
Mysterious land bridges were conjured out of nowhere
to account for the movement of plants and animals from
one continent to another. In the majority of cases, no
trace has ever been found of sunken land which could in
any way have fulfilled such a purpose. Thus, despite
some very good circumstantial evidence, the concept of
Figure 2.5. Simplified plots of continental movement shown by
changes in the direction of magnetic north.
continental drift went out of favour and remained so
until the late 1960s when it was once again revived. A
decade later there was virtually complete agreement
that continents had moved and that the findings of
Wegener concerning the climatic, biological, geological
and physiographical characteristics were correct.The
convictions and faith of a few scientists, mainly in the
southern hemisphere, were finally vindicated in the new
approach which came to be called 'plate tectonics'. The
evidence which made this new approach possible is
considered in the next section.
Plate tectonics
The possibility of continental mobility received fresh
impetus from a number of important discoveries made
during the two decades, 1956 to 1976. Significantly,
these discoveries were made by the very group of
scientists (geophysicists) who formerly had doubted the
theory of continental drift. Realisation of the significance
of residual rock magnetism and the implications
of mid-oceanic ridges did not appear separately to have
immediate importance for continental mobility. When
taken together, however, these concepts provided the
opportunity for a complete reappraisal of the ideas of
continental drift and for the first time gave a credible
means of demonstrating continental motion.
Firstly, a paper published by a group of British
scientists in 1956 revealed that the direction of magnetisation
of rocks could be related to the position of the
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Plate tectonics 17
Mid-Oceanic Ridge
, Volcanic Island
Continental Slope
and Shelf
. \South America
Volcanic rocks covered
with Red Clay and Ooze
Central Rift
(Elevated ridge
many earthquakes)
Figure 2.6. Diagrammatic cross-section of the Atlantic Ocean
indicating the mid-oceanic ridge.
magnetic pole of the Earth at the time the rocks were
laid down. When rocks cool from a magma, and even
when sedimentary rocks are deposited, the iron minerals
within them become slightly magnetised and their
polarity is in sympathy with the Earth's magnetic field
at the time and place of formation. By careful measurement
it is possible to reconstruct the position of the
magnetic pole at different stages of the past history of
the Earth. When this was done, it was found that the
results were consistent for any one continent, indicating
that either the pole or the continent, or both, could have
moved. However, when the previous positions of the
magnetic pole were plotted for other continents, differences
became apparent which can only be explained by
a differential movement of the landmasses (Figure 2.5).
A second important hypothesis was proposed by an
American, H.H. Hess in 1960, following the realisation
that the mid-oceanic ridges were a characteristic feature
of not only the Atlantic, but the other oceans as well. It
was appreciated for the first time, that the largest and
longest mountain range of the Earth's surface lay at the
bottom of the sea (Figure 2.6). It possesses peaks which
rise from the ocean depths to more than 4000 m above
sea level, making a total height greater than Everest, and
its length is more than 64000 km. As this enormous
feature of the oceanic floor is composed entirely of
volcanic outpourings, Hess suggested that the sea floor
cracked open along the line ofthe mid-oceanic ridge and
that new volcanic material came to the surface and
gradually spread on either side of the spreading centre.
This proposal was attractive as it offered a means of
solving the apparent youthfulness of the ocean floors,
virtually all of which are post-Cretaceous in age, and
explained at the same time why there was a lack of
sediment covering much of the sea floor.
Thirdly, during the period 1965 to 1975 there was the
realisation that the magnetism in the rocks on the sea
floor could be related to the age of the lava and that
together these three discoveries could be used to give
the most convincing proof yet available that the continental
plates had moved and are still moving. As Figure
2.7 shows, the reversals in the Earth's magnetic field are
faithfully recorded in the bands of lava extruded from
the spreading centre. Each side is pushed further and
further apart by the continuing process oflava extrusion
at a rate of between 1 and 9 mm per year. The original
continental drift theory assumed that the lighter granitic
or sialic rocks were floating upon the sima or denser
basaltic rocks of the sea floor. The modern interpretation
is that the continents are firmly embedded in the
sea-floor material which is as rigid and brittle as the
continental material. Both continental and sea-floor
materials are being moved along together on a large
fragment of the Earth's crust which is referred to as a
lithospheric plate. Each of these plates is between 150
and 200 km thick and is bounded by a margin which is
marked by localised seismic, volcanic and tectonic
activity (Figures 2.8 and 2.9). If new material is constantly
being introduced along the line of the midoceanic
ridges and spreading out from them, it follows
that it will meet material from another spreading centre.
It will either come into direct collison with it, it will
override it, or it will itself be overridden (Figure 2.10).
Three types of plate boundary are recognised:
1. At the mid-ocean ridge margin of a lithospheric
plate new material is being injected into the centre of the
spreading zone as the plates on either side move apart this
is known as a constructive margin.
2. Where island arcs, trenches and fold mountain
belts occur there is a convergent movement of the plates
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
18 Continental drift and plate tectonics
Gilbert i Gauss i Matuyama i Brunhesi Matuyama i Gauss i Gilbert
Cooling and Magnetisation
Figure 2.7. As new sea-floor material is produced at the spreading
centre, it takes the polarity prevailing when it cools. The names
Brunhes (normal), Matuyama (reversed), Gauss (normal) and
Gilbert (reversed) refer to periods when polarity alternated during
the past 4 million years.
Spreading ridge offset by transform faults
Collision zone
Subduction zone
Figure 2.8. The major lithospheric plates of the Earth.
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Plate tectonics 19
Figure 2.9. A map showing the location of major earthquakes
outlines the major lithospheric plates.
Destructive
margin
Conservative Constructive Destructive
margin margin
A S T HE^N^OSPHE
M E S O S P H E R E
Lithosphere
Figure 2.10. Types of plate boundary.
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
20 Continental drift and plate tectonics
and the margin of one plate is being pushed down
(subducted) into the asthenosphere - this is known as a
destructive plate margin.
3. Where the plates are moving past each other, as along
the famous San Andreas fault, and other transform
faults which cross the spreading zones of the midoceanic
ridges, lithospheric material is neither being
created nor destroyed and so these margins have been
called conservative margins.
The margins of the Atlantic Ocean were formed as
the North American and African/European plates
moved apart. Hence the name Atlantic or trailing
margins are used. Around the Pacific the margins are
mainly destructive with the plates moving together.
These are referred to as Pacific or collision margins.
These names simply describe a number of different
situations where the lithospheric plates interact. As the
ocean-floor material is different from the material of the
continents, it must be expected that the interaction of
the plates would be different:
Simple subduetion: the simplest case occurs where
oceanic material meets continental material, as is taking
place in the Andes. As continental material is so light,
the denser, oceanic Pacific plate is being subducted
beneath the South American plate where it is melted as
it descends into the asthenosphere, so providing magma
for volcanic activity.
Island arcs: these are a characteristic of the western
Pacific Ocean and reflect a complex situation which is
not altogether understood. A deep sea trench lies
alongside a single or double arc of islands, the inner arc
usually being volcanic. Island arcs differ in complexity
for they are not all backed by continental materials as
occurs in the Indonesian arc. In some cases oceanic
material is in collision with oceanic material, as occurs
in the South Sandwich arc of the south Atlantic. In the
Philippines it is thought two arcs have collided to give
the particular structures of that group of islands.
Continent to continent: where the Indian plate meets
the Asian plate the Indian plate has been thrust beneath
the Asian plate to give a double thickness of continental
crust. A similar, but not so spectacular situation occurs
in the mountains of Armenia where the Arabian plate is
thrust beneath the Eurasian plate.
Within the lithospheric plates lines of weakness are
indicated by a pattern of rifting seen in all continents,
but particularly in the rift valleys of east Africa. In the
past, fragmentation of Gondwana also seemed to have
taken place, by rifting along similar lines to the east
African rift valley with typical Y- or triple-junctions.
This rifting was preceded by intrusion of alkaline
volcanics which caused the doming of the area, and the
extrusion of basaltic lava flows. It is uncertain whether
the African rift valleys are a failed spreading centre. In
many cases two arms ofthe triple junction are successful
in developing a spreading centre and one arm fails; this
failed arm is known as an aulacogen.
In the process of continental growth, the results of
successive mountain-building episodes (orogens) have
been attached to the older cratons of basement complex.
In some cases materials of different ages are very firmly
attached, but in other cases the line of weakness
between them continues to be the seat of earthquakes.
Although the theory of plate tectonics has been
widely accepted in the earth sciences and it provides
answers to many of the problems of global tectonics, it
still does not answer all of the questions. One such
problem is the disparity between the length of the
spreading centres compared with the length of the
trenches where material is consumed. Similarly the
pattern of earthquakes is not always consistent with the
ideas of subduction, and it is apparent from the North
American continent that it is riding over the margins of
the Pacific plate regardless of the underlying structure.
The link between mountain building and plate tectonic
theory is not strong and the idea of compressive folding
is losing ground to ideas of gravity sliding in the
production of nappes and other folded structures seen
in Alpine-like mountains. The simple model of a
limited number of large lithospheric plates is breaking
down as it is found that there are many micro-plates
which are equally important in the determination of the
large-scale morphology of the Earth.
Up to this point in the discussion of plate tectonics
the size of the Earth has been assumed to remain
constant. However, this may not be a correct assumption.
It has been suggested that many of the problems of
interpretation could be solved if the continents could be
arranged on a smaller globe. The pattern of spreading
outwards from the Atlantic, Indian and Southern
oceans on an Earth of constant size suggests that much
more compression should exist around the Pacific
Ocean. In fact behind some of the island arcs, back arc
basins have subsidiary spreading centres showing that
the Pacific, too, has expanded. Although virtually
impossible to calculate, it appears that more new sea
floor is being created at spreading centres than is being
consumed in the subduction zones. The roughly polyDownloaded
from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Origin of the sea floor 21
Figure 2.11. Features of the ocean floor.
gonal shape of the lithospheric plates is consistent with
an expansion of the earth and, except for Antarctica, all
the other plates have migrated northwards from the
Southern Ocean spreading centre, again giving support
to the idea of an expanding Earth. With these cautionary
words, a consideration of the origin and development
of the ocean floors and the continental areas
follows in the next section.
Origin of the sea floor
It has been demonstrated that ocean-floor rocks are
different from the continental rocks. In general, this
difference can be expressed by describing the sea-floor
rocks as basaltic and the continental rocks as granitic.
The geological origin and development of sea floor and
continents is markedly different, so it is convenient to
discuss them separately.
Recent discoveries have shown that the morphology
of the oceanfloorreflects a dynamic state and not a static
situation. As the role of the seafloorin the broad pattern
of the Earth's morphology is seen now to be of basic
importance, it is clearly the most suitable place to begin
a study of world geomorphology. There are three major
elements in sea-floor morphology: the spreading
centres; the ocean basins; and the trenches. Each of
these features plays an important part in the understanding
of the global geomorphological development.
The spreading centres
The Atlantic, Indian and Southern oceans are characterised
by mid-oceanic ridges which mark the location
of spreading centres (Figure 2.11). The Pacific Ocean is
slightly different as the line of the spreading centre
swings from the south-eastern Pacific and eventually
passes on to land in California as the San Andreas fault,
eventually passing back to the ocean floor again off the
coast of north-western USA. The only other major land
area where the spreading centre emerges is in Iceland
which lies astride the mid-Atlantic ridge.
As revealed by surveys, the form of the mid-oceanic
ridge is of parallel strips of material extruded on either
side of the spreading centre. The overall height of the
mid-oceanic ridge depends partly upon the rate at
which material is being extruded and partly upon the
rate of spreading because away from the spreading
centre the newly extruded sea-floor material settles back
into the Earth's crust. If spreading is rapid, the midoceanic
ridge will be broad with gently sloping sides,
but if spreading is slow, the slopes of the ridge are much
steeper, as in the South Atlantic. In some cases it has
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
22 Continental drift and plate tectonics
Spreading Centre
Sea Mounts or
Guyots
Transform fault
'quake activity'
Reversals of magnetisms are mirror images
of those on other side of spreading centre
Figure 2.12. A spreading centre with a transform fault.
Elevated ridge with central trough,
intensive seismic activity,
extrusion of new sea floor
been possible to distinguish a rift-like valley between
the two sides of the spreading centre; in Iceland it is 45
km wide, and in the Indian Ocean the central rift valley
is 300 m deeper than the ocean floor on either side.
The mid-oceanic ridge is not continuous but comprises
a succession ofsegments where it has been shifted
out of line by transform faults oriented at right angles to
the ridge. As Figure 2.12 shows, this can lead to
situations where adjacent segments of the Earth's crust
are moving in opposite directions. Where this occurs
earthquakes are particularly common and they are often
accompanied by volcanic activity. Occasionally, the
amount of volcanic outpourings is sufficient to raise the
level of the lava above sea level to form one or more
volcanic islands. It is noticeable that islands further
from the spreading centre are larger than those close by.
This fits the general picture because the oldest volcanoes
have obviously continued to grow as they have
produced much greater quantities of lava over the
longer period of their existence. Most volcanoes cease to
be active after a period of 20 to 30 million years and only
a few, such as the Canary Islands, remain active for up
to 100 million years.
Extinct volcanoes on the sea floor are referred to as
guyots or seamounts, many of these do not grow
sufficiently to appear above the sea surface as islands
and even some of those which did become islands later
sank below the sea again as the crust settled away from
the spreading centre. Often guyots are flat-topped as
wave action has trimmed their summits. As they gradually
sink out of sight these guyots sometimes form the
foundations of atolls. The guyots are carried along on
the tectonic plate and they are ultimately consumed
when the edge of the plate passes into a subduction
zone. One such guyot has been discovered on the side of
the Tonga trench, appropriately tipped over as it is
being carried downwards beforefinallydisappearing for
ever; another sits on the floor of the Aleutian trench,
about to be consumed.
The ocean basins
Away from the spreading centres, the ocean floor
gradually descends to the level of the ocean basin floor,
alternatively called the abyssal plain. This descent from
the crest of the mid-oceanic ridge, at about 2000 m, to
the abyssal plain at 5000 m takes place in a series of steps
associated with successive outpourings of new material
from the spreading centre. As new material is intruded
along the line of the spreading centre, the previous
intrusions are pushed further and further apart. As each
intrusion became magnetised in sympathy with the
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Origin of the sea floor 23
i ^ \ \ 70
\\\\\\\\\\50\
50x
\ \ \ \ \ \ \ \ \
Figure 2.13. The mid-Atlantic ridge with parallel strips of new sea
floor. The age of these strips on either side of the spreading centre
is given in millions of years.
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
24 Continental drift and plate tectonics
Idealised Section through
Oceanic Crust
km
0 - .
1 0 -
20-I
Sea
Sediments
Basaltic pillow lava
Dolerite intrusive
complex
Gabbro-Basaltic layer
Mohorovicic discontinuity
Upper Mantle
Figure 2.14. Idealised section through the oceanic crust.
Earth's contemporary magnetic field, it follows that the
changes in polarity observed in the rocks are related to
changes in the Earth's magnetic characteristics.
Surveys have shown that this theoretical picture is
correct, and that the seaflooris made up ofa sequence of
parallel strips of rock some of which have reversed
magnetism. As polarity has changed many times in the
past and at varying intervals, it is possible to compare
the pattern of alternately magnetised strips from different
oceans. On all ocean floors the evidence is the same,
areas of new sea floor are symmetrical and the alternately
polarised bands are found to agree in age and
direction of magnetisation (Figure 2.13).
The floor of the ocean basins appears to have three
major layers. Overlying the mantle and the asthenosphere
(or weak plastic layer) is the 'oceanic layer'
composed of a basic igneous rock complex with many
doleritic dykes passing downwards into layered gabbros,
making a layer 5 km thick. Upon this is a second
layer, 1.5 to 2.0 km in thickness, composed of basaltic
pillow lavas which have been extruded under water and
rapidly cooled. Adjacent to continental areas, this,
volcanic layer may become interstratified with consolidated
sedimentary rocks such as limestone and shales
(Figure 2.14).
The trenches
If new sea-floor material is constantly being intruded
along the lines of the spreading centres and the size of
the Earth remains the same, it follows that an approximately
equal amount of material must be destroyed or
absorbed back into the deeper layers of the Earth's
crust. This takes place in the deep oceanic trenches.
In the nineteenth century a British oceanographic
survey vessel found depths greater than 8000 m (24 000
ft) near Tonga in the Pacific Ocean. Subsequently, it
was found that the deepest parts of this abyss was 12 000
m (35 000 ft) deep and that it occurred in a long narrow
V-shaped chasm where the sea was approximately seven
times as deep as the Grand Canyon. Other trenches
were subsequently found and in most cases they were
found to be slightly over 12000 m deep. Characteristically,
these deepest parts of the oceans were not in the
middle but around the periphery, near the Aleutian
Islands, Japan, the Marianas Islands, the Philippine
Islands, Java, Peru, Chile and Guatemala (Figure 2.15).
Significantly, there is no trench off the western coast of
the United States, but the structures associated with the
San Andreas fault indicate a completely different geological
situation. In the Atlantic Ocean, trenches are not
so common, only being found in association with the
island arcs of the West Indies and the South Sandwich
Islands. The Indian Ocean has a trench south-west of
Sumatra and extending towards the Andaman and
Nicobar Islands.
The trenches are associated with negative anomalies
of gravity and with earthquakes and volcanic activity on
their landward side. The amount of heat flowing from
the interior of the Earth in these regions is less than
normal, suggesting that there might be a downward
movement ofcool rock which would reduce the outward
flow ofheat. As these trenches are the deepest part of the
ocean, it might be thought that they would have an
accumulation of sediment in them, but this is not so;
only when trenches cease to be active do they accumulate
sediments to any extent. It is not thought, therefore,
that trenches develop into geosynclines, as the latter are
characterised by large sediment accumulations.
The active role of the deep sea trenches concerns the
leading edges of the tectonic plates. Previously it has
been described how new sea floor is created at the
spreading centres; it is also obvious that an equal
amount must be removed in some way. As a crustal
plate grows, its leading edge is destroyed at the same
rate by reabsorbtion into the deeper layers of the crust.
Where tectonic plates are moving together at speeds of
more than 5 cm per year, one plate normally slides
beneath the other and is consumed in a subduction zone
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Origin of the continental areas
Asia
Acapulco-Guatemala Trench
South
America
• Deep centres of earthquake activity
Trench
Figure 2.15. Trenches of the Pacific and location of deep
earthquake foci. (After Fisher and Reville, 1955.)
(also known as a Benioff zone). The leading edge of the
plate passes downwards into the asthenosphere and this
zone is usually marked by many earthquakes and by
intense volcanic activity (Figure 2.16). At lower speeds
of approach it is possible for two colliding plates to
buckle at the edges and produce fold mountain ranges
between the two advancing plates.
Origin of the continental areas
Uniformitarianism is a principle of geology which, put
simply, states that the present is the key to the past.
Observation of present-day processes leads to an understanding
of what has happened in geological history
throughout which a similar set of processes has operated.
It is quite logical that if erosion takes place in one
locality, there must be a corresponding deposition of
material elsewhere.
For many years now, it has been understood that
eventually all eroded material becomes deposited in
downward-sagging belts of the Earth's crust known as
geosynclines. This name was first given to these features
by Dana in 1873. At first it was thought that the weight
of sediment would initiate a downwarping, but this is
now thought not to be the case and an independent
downwarping occurs first, only to be reinforced later by
the weight of accumulating sediment. The downward
movement of the crust to accommodate the additional
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
26 Continental drift and plate tectonics
/ss/ s / / / / / / / / / / / / / / / / /
Lithosphere Plate ////////
\Subduction zone •
- v \
Upper Mantle
Figure 2.16. Section of deep sea trench with subduction zone and
location of earthquake foci.
load of sediment is relatively slow. It has been calculated
that 12000 m of sediment accumulated between
the Cambrian and the Permian in the Appalachian
region and that this represented accumulation at the
rate of 10 cm in 2500 years on average. It was assumed
that as the downward-sagging sediments reached
deeper and deeper into the Earth's crust they would be
subject to pressure and heat which would produce
metamorphic rocks such as schists and gneisses.
Finally, movement of the forelands on either side of the
geosyncline brought compression of the geosyncline
resulting in uplift, folding and faulting. Igneous activity
accompanied this activity with volcanoes and the
instrusion of sills and dykes. In the deeper parts of the
geosyncline plutonic intrusion invaded the sedimentary
rocks to give granitic cores beneath the mountain
ranges. These were only revealed subsequently when
the mountains were eroded down to their roots.
Thrust-faulting or nappes (and less complicated
folding) are greatest on either side of the geosyncline
whereas in the centre considerable median masses may
be uplifted, or in some cases downfaulted (Figure 2.17).
The Tibetan plateau and the Hungarian basin are
examples of uplifted/downfaulted but relatively
unaffected land masses which were affected by the
Alpine orogeny. The Tyrhennian Sea is an example ofa
large downfaulted basin. The ideal picture of a geosyncline
given in many textbooks suggests a symmetrical
upthrusting resulting from an inward movement of the
forelands on each side. However, these movement are
not always evenly matched. In the case of the Tethys
geosyncline it appears that the African plate moved
against the European foreland over which the northern
folds have been thrust.
Although the concept ofplate tectonics has revolutionised
our understanding of the ocean floor, it is perhaps
not quite so obvious how the theory can be related to the
origin and growth of the continental areas. Fortunately,
many aspects of the geosynclinal theory previously
described can be linked to the ideas of plate tectonics to
give a satisfactory explanation of the growth of continental
masses.
Radioactive dating of the basement rocks indicates
that there are nuclei or core areas in the continents
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Origin of the continental areas 27
Figure 2.17. Diagram of a mediean mass (zweisengebirge.)
km
10Granitic
Layer 20-
30-
40Basaltic
Layer
50-
60Upper
Mantle
Granodiorite
Enderbite
Conrad Discontinuity
Amphibolite
Gabbros
Eclogite
Mohorovicic discontinuity
Peridotite
Pyrolite
Figure 2.18. Idealised section through continental crust.
which are more than 2000 million years old, around and
upon which successively younger deposits have accumulated.
The landmasses have a different structure and
composition from the ocean floor, already described.
Overlying the mantle and the Mohorovicic discontinuity
is a layer of basaltic composition similar to that
found beneath the oceans (Figure 2.18). It has been
described in Chapter 1 how this layer crops out only at a
small number of locations and it appears to be formed of
amphibolite, a metamorphosed form of basalt. Normally,
it lies between 35 and 50 km below the surface
except for the few fragments which have been brought
to the surface in the process of mountain building.
The upper part of the continental crust lies above the
Conrad discontinuity; it is composed of granitic rocks
which extend to a depth of 35 km. These rocks are
mCratons - stable at
least since ca.1 500 m.y.
ago
Orogens - deformed
during the the past ca.
1200 m.y.
Figure 2.19. Cratons of Africa.
comprised of 92% igneous and metamorphic rocks with
only 8% sedimentary material such as limestones,
sandstones and shales, which lie near to the surface. The
continental material lying above the Conrad discontinuity
is formed from a combination of structurally
complex, older shield areas, bordered by younger,
intensively-folded sedimentary rocks.
Approximately 80% of the continental areas are
underlain by the older, Pre-Cambrian rocks, which are
called the basement complex. These oldest parts of the
continents have grown by accretion as tectonic activity
metamorphosed former sediments into crystalline
rocks. Growth did not take place uniformly, but in
phases associated with mountain building and igneous
activity, which reached peaks around 2700, 1800 and
1000 million years ago. Most of the rocks of these older
shield areas have experienced pressures and temperatures
unknown in the younger fold mountains. The
approximate ages of different parts (cratons) of the
basement complex of Africa are shown in Figure 2.19.
Some parts of the basement complex are covered
extensively with undeformed or slightly deformed sedimentary
rocks. Others have accumulated in deep
persistent downwarps which extend to 15 km depth
below the surface. Where the Pre-Cambrian basement
complex is exposed at the surface, it can be seen to be
both structurally and chemically complex. Intense foldDownloaded
from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
28 Continental drift and plate tectonics
X X X X X
X X X X
X X X X
X X X X
X X X >1
X X X
X X X XI
X X X
Blue Ridge
Allegheny
Front
Ridge and Valley
New Coastal Plain
Shelf
Figure 2.20. Eugeosyncline and miogeosyncline couplet.
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016
Origin of the continental areas 29
ing is characteristic and may be associated with acid or
basic igneous rocks, the latter possibly being former
pieces of the oceanic floor caught up in orogenesis.
Rocks from the basement complex, too, can be caught
up in later orogenic activity and may form significant
parts of the roots of mountain ranges.
An earlier view of orogenesis was given at the
beginning of this section in the description of a geosyncline.
The theory left many points unanswered which
the recent discoveries associated with plate tectonics
helped to answer. Recent observations have led to the
suggestion that there are two parts of a geosyncline (a
geosynclinal couplet), lying side by side and each
playing a distinct role (Figure 2.20). According to Deitz
(1972) deposition of shallow-water beds occurs in a
miogeosyncline (lesser geosyncline) where crustal
downwarping occurs at the edge of a continental mass.
These sediments increase in thickness seawards and end
at the edge of the continental shelf. Alongside the
continental margin is the eugeosyncline (true geosyncline),
comprising a wedge of sediments accumulated
on the oceanfloorand embanked against the continental
rise. The sediments of the eugeosyncline are derived
from muddy suspensions, known as trubidites, which
have the appearance of thinly-bedded, poorly-sorted
sands and silts interbedded with clays. Such sedimentary
rocks have also been referred to as graywacke or
flysch.
One of the most likely places for a geosynclinal
couplet would be on the trailing edge of a lithospheric
plate. It is suggested that as new sea floor was intruded
an initial sagging took place into which the deposit of
the eugeosyncline began to accumulate. Once started,
subsidence continued under the weight of the sediments
as isostatic compensation took place. On the
landward side, this downward flexing provided the
continuing shallow water conditions observed in the
miogeosyncline. Altogether it is thought that a depth of
up to 15 km of sediment could accumulate.
As the ocean basins are not fixed, but continually
altering according to the lithospheric plate, it is possible
that a subduction zone could be initiated by the deeply
sagging eugeosyncline. Then as the ocean floor
advanced towards the landmass, the sedimentary content
of the eugeosyncline would be crumpled against it
as well as being thrust deeply into the Earth's crust.
Isostatic readjustment would eventually follow to raise
the folded material as a mountain mass.
Alternatively, the eugeosyncline could be squeezed
between two approaching continental masses. Once
again crumpling would occur and the sediments would
be strongly compressed, leading to the production of
metamorphic rocks and to a thickening of the upper
layer of the crust. Isostatic readjustment would then
come into operation as in the previous case, associated
with volcanic intrusion and emplacement of batholiths
in the core of the folded sediments.
The miogeosyncline with its sedimentary deposits
would be less strongly compressed and only affected by
volcanic intrusions. The degree of folding would be
greatest at the seaward edge of the miogeosyncline and
would decrease inland until unfolded beds mantled the
basement rocks formed earlier. As compression and
mountain building took place, contemporaneous continental
deposits of conglomerates, sandstones and
shales, known collectively as molasse, accumulated to
cover partially the slightly folded miogeosyncline beds
as erosion attacked the rising mountain chain.
The theory of plate tectonics proposes that the
surface of the Earth is composed of eight major lithospheric
plates and a number of smaller areas, known as
micro-plates. The major plates are the Eurasian, African,
Indo-Australian, North American, South American,
Pacific, Nazca and Antarctic plates. The microplates
include smaller sub-continental areas such as
occur in the Caribbean, south Pacific and Anatolian
areas.
This account of world geomorphology attempts to
discuss the geomorphology of the lithospheric plates
which form the primary sub-divisions of the Earth's
morphology. Each major plate has a submarine and a
terrestrial component. Knowledge of the terrestrial
geomorphology of each major plate varies greatly in
detail; of the submarine geomorphology, which is less
easily observed, only the briefest outlines are known.
Downloaded from Cambridge Books Online by IP 147.251.65.55 on Fri Mar 11 09:32:53 GMT 2016.
http://dx.doi.org/10.1017/CBO9781139170154.003
Cambridge Books Online © Cambridge University Press, 2016