Not to be cited without prior reference to the author
ICES 2008 CM/C:05
Are seamounts like oceanic islands for fish diversity and speciation?
Paul J B Hart
Department of Biology
University of Leicester
Leicester LE1 7RH UK
Tel: 44 (0)116 2523348
Fax: 44 (0)116 2523330
Email: pbh@le.ac.uk
Abstract
Seamounts can be considered as islands in the deep. For many species, depth
is just as much a barrier to dispersal as is the water between islands such as
along the Hawaiian chain. This leads to the hypothesis that seamounts could
be places where speciation leads to many new forms. Estimating the number
of endemic species on seamounts is fraught with difficulties caused mostly by
remoteness from research centres, difficulty of access due to depth and
sampling problems deriving from the nature of the habitat. A review of all
seamount studies from all oceans estimated that 11.6% of fish on seamounts
are endemic. Recent work on the Norfolk Ridge, the Lord Howe, and the
Tasmanian seamounts and the Sala-y-Gómez chain of seamounts in the
Pacific have found that potential endemics might form 29-51% of fish caught.
Although based on only two groups of seamounts, these proportions for
endemism support the view that seamounts might be the location of active
speciation. Using recent theoretical advances in island biogeography and
community dynamics, a conceptual framework is offered that could help steer
future research on seamount diversity and speciation. The major hypothesis is
that seamounts can be regarded as equivalent to oceanic islands so that the
same theory used to predict patterns of island diversity can be applied to
seamounts. Although this approach cannot ease the acquisition of data from
seamounts it could help to direct sampling effort to maximise the value of
expensively gathered data.
Introduction
Documenting the existence and origin of biodiversity in the world’s oceans is
made difficult by the nature of the habitat. Despite this the task cannot be
neglected given that the oceans cover two thirds of the world’s oceans and is
where a significant amount of food for human consumption is produced. We
need to know what is in the sea and how diversity arises if management and
conservation are to be effective. Seamounts are numerous, with perhaps
100,000 existing throughout the World’s oceans (Kinchingman et al 2007), and
1
Not to be cited without prior reference to the author
they are analogous to oceanic islands but embedded entirely in the aquatic
realm. The fauna and flora of oceanic islands are of great interest to those who
want to understand the origin of biodiversity, and endemism is a
characteristic of many of these islands (Whittaker and Fernández-Palacios,
2007; Grant & Grant, 2008; Price, 2008). As seamounts share many
characteristics with oceanic islands they could be sites for high levels of
endemism. In this paper, I plan to examine the evidence for this hypothesis as
it relates to the fish fauna and to outline the data that would be required to
test it. At this stage, so little is known about the fish on seamounts, that
definitive tests of ideas are as yet out of reach.
A second hypothesis that will be used to aid thinking is that birds are the
terrestrial equivalent of fish in the marine environment. There are clear
differences between the two groups, the main one of which is the size,
number and mobility of eggs and young. This difference has significant
consequences for distribution but in many other respects the two groups live
in a medium that presents the organism with similar problems. Both groups
contain organisms the populations of which are distributed on a similar
spatial scale between 102 to 104 km (Dawson and Hamner, 2008), and
organisms in both have to contend with moving in a fluid medium.
Seamounts as underwater islands
As with many oceanic islands, seamounts are most often the result of volcanic
activity. The volcano was just not big enough to push material above sea
level. Some seamounts are also the remnants of volcanic oceanic islands that
have been destroyed by erosion (Price and Clague, 2002). Volcanic islands can
reach significant altitudes above sea level but during the period immediately
after the eruption the island begins to sink as the weight of material depresses
the earth’s crust (Price and Clague, 2002; Wessel, 2007). The same will happen
with volcanoes that do not break the sea surface but still achieve a great
height above the sea floor. Seamounts that are 2-3,000 meters are not
uncommon. This gradual sinking is likely to be as important for speciation on
seamounts as it is on islands. The rapid period of sinking can last for about 1
Ma and can reduce the height of the seamount by 1000 to 1500 m (Price and
Clague, 2002).
Whilst an island exists there is a sharp demarcation between the terrestrial
and marine habitats. This has obvious consequences for the biota in that it sets
a limit to their distribution. At first consideration, a seamount does not
appear to have these sharp limits between one habitat type and the next but
this perception is more a result of our land based viewpoint than the
biological reality. Marine organisms are just as limited in the conditions they
2
Not to be cited without prior reference to the author
can tolerate as are terrestrial plants and animals and conditions in the sea
change rapidly with depth (Mann and Lazier, 2006, Ch 3). For a start, light
only penetrates to at most around 200 m which restricts productivity to the
surface layers of the ocean. Below 1000 m any organism has to obtain its food
either from other denizens of the deep or from organic material falling from
the euphotic zone. For these reasons, a seamount that has it’s top 200 m in the
euphotic zone will be a very different habitat from one that has its peak at 800
m depth. For the purposes of this paper, seamounts that project into the
euphotic zone only will be considered, as it is most likely that only these
seamounts have biodiversity that is high and dynamic. The 500 m depth mark
will be considered equivalent to the sea/land interface characteristic of an
oceanic island. In this sense, the tops of seamounts projecting into this zone
can be thought of as islands (Figure 1). Taking the analysis one stage further it
could be said that there is a continuum from a volcanic structure that has its
top projecting beyond the sea surface to a low hump near the sea floor with a
height of only a few hundred metres. This continuum will have a gradation of
conditions for life with two sharp boundaries; the split between the gaseous
and liquid fluid bathing the flanks of the volcano and the boundary between
the illuminated and dark region of the liquid realm.
Spatially the biota on these subsurface islands will be subject to the same
factors that influence the biota on terrestrial islands in the sea. They will be
separate from other islands and from continental shelves in the euphotic zone
and they will be exposed to immigration of organisms from these other
locations. Equally, some of these seamounts will be expected to gradually sink
as time passes so that eventually their tops pass under the 500 m mark and
out of the direct influence of surface events. My analysis applies to these
seamount tops, during their existence as islands in the euphotic zone.
Theories of island biogeography
The first, and still the most cited, theory which attempts to explain the
number of species found on islands is by MacArthur and Wilson (1967). Their
hypothesis was that the number of species on an island was determined by
the balance between species arriving by immigration and species becoming
extinct. The basics of their theory are shown in Figure 2, now a textbook
figure. Elaborations on this theory have been numerous but the work of
Hubbell (2001) and most recently Whittaker et al., (2008) have taken the
theory several steps further. This paper is not the place for a full explanation
of this new work, but a brief outline is required to make the rest of the paper
understandable.
Hubbell’s (2001) theory and that proposed by Whittaker et al., (2008) are
related to each other and both derive from the MacArthur and Wilson’s (1967)
theory. The latter is a neutral theory meaning that the rates of immigration to
an island and extinction of species on it are the same for all species. This
3
Not to be cited without prior reference to the author
simplification means that no species is better or worse at getting to an island.
The additional variables that influence the number of species on an island are
the area of the island and how far it is from the mainland or other source of
species.
Hubbell (2001) also proposes that immigration and extinction are the same
but he applies this to a different level or organisation proposing that the
neutrality applies to individuals rather than to species. A further elaboration
is to make explicit something that was inherent in MacArthur and Wilson’s
(1967) theory but not dealt with properly and that is that it assumed that
communities are made up of an eclectic mix of species largely thrown
together at random. Hubbell (2001) calls this the dispersal-assembly
perspective. This is in opposition to the niche-assembly perspective which
argues that the species in a community have been moulded by natural
selection to fit a set of niches and that competition and character displacement
are the two main mechanisms bringing this about. The niche-assembly
hypothesis underlies the process of adaptive radiation which has competition
for resources as a main driving force (Schluter 2000).
The theory proposed by Hubbell (2001) is restricted to a very specific
definition of a community. In his terms an ecological community is ‘… a
group of trophically similar, sympatric species that actually or potentially
compete in a local area for the same or similar resources’ (p 5). So, for fish this
would mean all algal grazers or all invertebrate feeders. The theory would not
apply to full ecosystems encompassing both predators and their prey.
Although Hubbell (2001) argues that this definition is not as restrictive as it
seems, it does lead to the possibility that the scope of his theory will be
misunderstood as his definition of a community is so restrictive.
The original theory of island biogeography did not say anything about the
evolution of species once they had arrived on an island. Evolution is part of
Hubbell’s (2001) development as it is of the theory proposed by Whittaker et
al. (2008). The latter is tailored to the particular histories that will be
experienced by an oceanic island which will exist for a limited period of time.
As I am assuming that seamount islands will also last for a limited period of
time, it might be expected that they too will be the scenes of speciation.
Whittaker et al’s (2008) theory of island biogeography is shown in Figure 3.
This theory is labelled as the General Dynamic Model or GDM. As the Figure
shows, immigration and extinction are still the important inputs and outputs
of the system but now as the island ages speciation rises to a peak and then
falls away as biodiversity increases and there are fewer opportunities for new
forms. The theory predicts that large remote islands will be dominated by
evolutionary change more than will islands closer to source landmasses. Note
that the carrying capacity of the island, labelled as K, increases to a peak and
then falls as the island is reduced in height and area by erosion and
subsidence.
4
Not to be cited without prior reference to the author
The GDM has random immigration and extinction events providing basic
inputs to an island but then assumes that competition for resources can drive
speciation on established islands. Animals or plants found as endemics on
only one island are taken as a measure of evolutionary dynamics within an
archipelago. It has been argued by Emerson and Kolm (2005 a, b) that there
is an association between high species numbers on an island and a high
proportion of single island endemics. These authors argue that high diversity
leads to increased competition for resources and to greater rates of speciation.
To test their GDM, Whittaker et al (2008) devise a regression equation, which
predicts that the diversity on an island a given distance from a source of
immigrants will be a functions of time and island area. This equation is
Diversity = a + b(Time) + c(Time2) + d(logArea)
where a, b, c and d are fitted parameters.
The predictive power of the equation is tested against data sets from the
Canary Islands, Hawaii, the Galápagos, the Marquesas and the Azores.
Generally, diversity in the data sets used is well described by the regression.
If applied to seamounts the GDM would lead to the expectation that species
diversity in a particular taxon would be a function of the age of the seamount,
its distance from a source of immigrants and its area. We would also expect to
find single seamount endemics although the proportion they form of all
endemics might be variable.
What information do we have on the diversity and levels of endemism in the
fish communities of seamounts?
Levels of diversity and rates of endemism on seamounts have been reviewed
by Stocks and Hart (2007). At present estimates of species diversity for any
group are beset by problems derived from sampling. When attempting to
compare seamount and continental margin diversity the latter will usually
have been sampled with greater intensity than will the seamounts. Also the
sampling gear is more likely to sample the relatively smooth bottom of the
shelf margin more effectively than it will the rough and rocky seamount.
Seamounts are also under sampled as shown by plots of numbers of species
collected against number of samples for 180 seamounts. This is always a
straight line for all seamounts sampled so far with no sign of the number of
species reaching an asymptote. This implies that we still do not have good
estimates of total species for any seamount.
5
Not to be cited without prior reference to the author
The data on fish is not yet good enough to examine whether continental
margins have more or less species that do seamounts. Equally, species
evenness, which provides and indication of whether or not the community is
dominated by a few species, cannot be properly estimated. The only
indication we have that evenness might be low, is the evidence from
commercial fisheries on seamounts, which tend to be dominated by one or a
few species.
Stocks and Hart (2007) defined endemism as a species that is found on one
seamount or a group of seamounts close together and nowhere else in the
ocean. It is never possible to be sure that a species is endemic, especially as
seamount sampling is so poor. There is always the chance that the apparently
endemic species exists elsewhere but has just not yet been discovered.
An early survey by Wilson and Kaufmann (1987) found that 11.6% of 449
species of fish from 100 seamounts were endemic. 72% of the data came from
just 5 seamounts, so this survey is hardly definitive. Richer de Forges et al.,
(2000) surveying the Norfolk Ridge, the Lord Howe seamounts and the
Tasmanian seamounts in the southwest Pacific found that 29-34% of all fishes
caught were new or potentially endemic. Earlier, Parin et al., (1997) had found
that 44% of 171 species of fish caught were endemic on the Nazca and Sala-yGómez chain of seamounts in the southeast Pacific. The survey of seamount
fishes by Froese and Sampang (2004) estimated that only 12% of fish from 60
seamounts were endemic.
An expectation dependent on endemism is that fish that have evolved on just
one seamount or a group of closely placed seamounts would be genetically
different from fish that are closely related by on distant seamounts. Evidence
reviewed by Stocks and Hart (2007) shows no support for genetic
separateness although the small amount of data available does not make this
a strong solution.
What do we know about the ages and areas of seamounts?
Both the theories of species diversity proposed by Hubbell (2001) and
Whittaker et al., (2008) incorporate speciation as a significant contributer to
the number of species that is achieved on an island. For oceanic islands, their
life-time determines the time span available for speciation to take place and
its rate is predicted to be at its highest some time after the origin of the island
and after sufficient immigrants have arrived to bring about interspecific
competition and the conditions for an adaptive radiation. On some islands,
such as the Galápagos, natural selection can bring about phenotypic change
very quickly (Grant and Grant, 2008) and in temperate lakes in British
Columbia, Canada, which can be regarded as freshwater islands in a ‘sea’ of
land, speciation in threespine sticklebacks (Gasterosteus aculeatus) has occurred
within a time span of around 10,000 years (Schluter and MacPhail, 1992).
6
Not to be cited without prior reference to the author
Mostly, speciation is given more time and oceanic islands are likely to last at
least a million years, assuming that they achieve a considerable size on first
origin (Price and Clague, 2002).
For seamounts we need to know how old they are and more significantly in
the context of this paper, how long does the seamount that starts life with it’s
peaks in the top 500 m of water remain there. Wessel (1997) estimates the ages
of 8,882 seamounts in the Pacific and shows that ages range from a few to
160+ million years old. This work is not specific enough for the current
approach as it does not attach ages to specific seamounts. A study that does is
Chaytor et al., (2007) who provide detailed bathymetry and age estimates for
14 seamounts in the Cobb and the Kodiak-Bowie seamount chains. Seamounts
in these chains originated from hotspots off the west coast of North America
and seamount age increases, with some exceptions, from east to west. The
youngest seamount is Bowie off Haida Gwai (Queen Charlotte Islands),
British Columbia, at less that 0.7 Ma, whilst the oldest is Patton close to the
Aleutian Islands at 29.7 Ma.
The seamounts in the Gulf of Alaska vary considerably in height and depth
below the surface (Table 1). By the definition used in this paper, most have
peaks that are below the euphotic zone. For those with peaks close to the
seafloor, it might be expected that most fish species will be of abyssal origin.
For the seamounts with peaks that are considerably above the seafloor but
nevertheless below the euphotic zone, island conditions could still hold and
there would be opportunities for speciation leading to endemism. What is
now required is a list of fish species found on each of these seamounts. The
database Seamountsonline (seamounts.sdsc.edu) has data for seven out of the
fourteen seamounts in Table 1 but there is only one record of single species of
a fish. Consequently nothing can be deduced about diversity or endemism.
From island biogeography it can be predicted that seamounts closest shelf
areas of an equivalent depth might have more species than those far out into
the Gulf of Alaska. A further prediction would be that seamounts with larger
areas, as indexed by summit area, would have more species than those with
small areas.
The relationship between species numbers, area and distance are predicted
from island biogeography and could mirror what is true of the numbers of
bird species on oceanic islands. As argued earlier, birds on islands can be
taken as analogues of fish on seamounts. On oceanic islands the number of
endemic species of bird is linearly related to the area of the island and the
distance the island is from the mainland source of immigrants (Price 2008).
Taking an evolutionary perspective, island bird diversity is also expected to
be greater on old islands that have existed long enough for adaptive
radiations to take place (Grant and Grant, 2008). For seamounts in the Gulf of
Alaska the oldest seamounts are also close to the mainland so it is on the
seamounts in the middle of the two chains (Denson, Welker and Pratt; Scott
7
Not to be cited without prior reference to the author
and Campbell) that one might expect to find the lowest fish diversity but
perhaps the highest levels of endemism. The GDM of Whittaker et al., (2008)
also predicts that on large remote islands the number of species will be
strongly influenced by speciation.
Discussion
This paper has proposed a conceptual and theoretical framework within
which the diversity of fish on seamounts can be analysed and discussed. All
we need now is good data!
The argument presented depends on the contention that within the sea there
are boundaries at different depths that define the patterns of ecological and
evolutionary processes. The chief boundary is between the euphotic and
abyssal zones, essentially a boundary defining habitats with and without
light. The lack of primary production in the abyssal zone defines a trophic
regime which is fundamentally different from the trophic opportunities that
exist in the eutrophic zone. As a result one might expect different selection
pressures to hold and this is born out by the very marked adaptations that
deep sea fish have to living in a food poor habitat (Marshall, 1971). These
differences mean that in any analysis of diversity on seamounts, depth must
be a prime variable considered, but within a given depth zone, I am claiming
that the same theory can be used to predict patterns of diversity. Within a
given depth zone, seamounts are equivalent to oceanic islands at the surface.
A further point to make is that studies of diversity need to take trophic level
into account. The main theories of diversity on oceanic islands, particularly
Hubbell’s (2001) neutral theory, apply only to organisms at the same trophic
level. The theory of Whittaker et al., (2008) is not specific about it’s trophic
specificity but as it proposes that competition on an increasingly populated
island will lead to speciation, the theory implies that it only treats species
living at the same trophic level.
The chief problem with any study of diversity and endemism on seamounts is
the shortage of data and the difficulty inherent in obtaining it. Seamounts are
often remote and hard to get to and it is expensive to carry out surveys at the
level of detail required. In addition, the sampling methods most often
employed are non-selective and lacking in discrimination with respect to
depth and precise habitat. Dredges and trawls just catch a collection of species
from between two depths, but there is little chance of obtaining more spatially
defined information on the species caught. Remote vehicles with cameras do
allow a more detailed survey of the spatial relations between organisms on a
seamount, but such equipment is expensive and can only sample small areas
as a time. We need good data from many seamounts within an oceanic region
before it will be possible to determine whether or not seamounts are
equivalent to oceanic islands in terms of their species diversity. The Gulf of
8
Not to be cited without prior reference to the author
Alaska seamounts discussed earlier could provide the information required if
sampled properly but even with this location, there are too few seamounts at
a given depth to test properly any hypotheses about the origins and
maintenance of diversity.
The thoughts expressed in this paper are just a start to the problem of
understanding what the level of species diversity is on seamounts and
whether seamounts are centres of speciation. The analysis presented is
unsatisfactory because there is so little good data making it impossible to test
the ideas proposed against the state of nature. From the point of view of
conservation and management of seamount fish, we should be thinking in
terms of closing all seamounts to exploitation until we understand their status
more fully. At present we have only a weak scientific basis on which to plan
conservation and management (Pitcher et al., 2007).
9
Not to be cited without prior reference to the author
References
Chaytor, J. D., Keller, R. A., Duncan, R. A. and Dziak, R. P. (2007) Seamount
morphology in the Bowie and Cobb hot spot trails, Gulf of Alaska.
Geochemistry, Geophysics, Geosystems, 8 (9) doi: 10.1029/2007GC001712.
Dawson, M. N. and Hamner, W. M. (2008) A biophysical perspective on
dispersal and the geography of evolution in marine and terrestrial systems.
Journal of the Royal Society Interface, 5, 135-150.
Emerson, B. C. and Kolm, N. (2005a) Species diversity can drive speciation.
Nature, 434, 1015-1017.
Emerson, B. C. and Kolm, N. (2005b) Emerson and Kolm reply. Nature,
doi:10.1038/nature04309.
Froese, R. and Sampang, A. (2004) Taxonomy and biology of seamount fishes,
In: Seamounts: Biodiversity and Fisheries (eds. Morato, T. and Pauly, D.).
Fisheries Centre Research Reports, 12 (5), 25-32.
Grant, P. R. and Grant, B. S. (2008) How and why species multiply. The radiation
of Darwin’s finches. Princeton University Press, Princeton and Oxford.
Hubbell, S. P. (2001) The unified neutral theory of biodiversity and biogeography.
Princeton University Press.
Kinchingman, A., Lai, S., Morato, T. and Pauly, D. (2007) How many
seamounts are there and where are they located? In Pitcher, T. J., Morato, T.,
Hart, P. J. B., Clark, M. R., Haggan, N. and Santos, R. S. (Eds) Seamounts:
ecology, fisheries and conservation. Blackwell Publishing, Oxford. Ch 2.
MacArthur, R. H. and Wilson, E. O. (1967) The theory of island biogeography.
Princeton University Press, Princeton.
Mann, K. H. and Lazier, J. R. N. (2006) Dynamics of marine ecosystems. 3rd
edition. Blackwell Publishing, Oxford.
Marshall, N. B. (1971) Explorations in the life of fishes. Harvard University Press,
Cambridge.
Parin, N. V., Mironov, A. N. and Nesis, K. N. (1997) Biology of the Nazca and
Sala y Gómez submarine ridges, an outpost of the Indo-West Pacific fauna in
the eastern pacific Ocean: composition and distribution of the fauna, its
communities and history. Advances in Marine Biology, 32, 145-242.
10
Not to be cited without prior reference to the author
Pitcher, T. J., Morato, T., Hart, P. J. B., Clark, M. R. Haggan, N. and Santos, R.
S. (2007) The depths of ignorance: an ecosystem evaluation framework for
seamount ecology, fisheries and conservation. In Pitcher, T. J., Morato, T.,
Hart, P. J. B., Clark, M. R., Haggan, N. and Santos, R. S. (Eds) Seamounts:
ecology, fisheries and conservation. Blackwell Publishing, Oxford. Ch 21.
Price, J. P. and Clague, D. A. (2002) How old is the Hawaiian biota? Geology
and phylogeny suggest recent divergence. Proceedings of the Royal Society,
London, 269, 2429-2435.
Price, T. (2008) Speciation in birds. Roberts and Co, Greenwood Village,
Colorado.
Richer de Forges, B., Koslow, J. A. and Poore, G. C. B. (2000) Diversity and
endemism of the benthic seamount fauna in the southwest Pacific. Nature,
405, 944-947.
Schluter, D. (2000) The ecology of adaptive radiation. Oxford University Press,
Oxford.
Schluter, D. and McPhail, J. D. (1992) Ecological character displacement and
speciation in sticklebacks. American Naturalist, 140, 85-108.
Stocks, K. I. and Hart, P. J. B. (2007) Biogeography and biodiversity of
seamounts. In Pitcher, T. J., Morato, T., Hart, P. J. B., Clark, M. R., Haggan, N.
and Santos, R. S. (Eds) Seamounts: ecology, fisheries and conservation. Blackwell
Publishing, Oxford. Ch 13.
Wessel, P. (1997) Sizes and ages of seamounts using remote sensing:
implications for intraplate volcanism. Science, 277, 802-805.
Wessel, P. (2007) Seamount characteristics. In Pitcher, T. J., Morato, T., Hart,
P. J. B., Clark, M. R., Haggan, N. and Santos, R. S. (Eds) Seamounts: ecology,
fisheries and conservation. Blackwell Publishing, Oxford. Ch 1.
Whittaker, R. J. and Fernández-Palacios, J. M. (2007) Island biogeography:
ecology, evolution, and conservation. 2nd Edition. Oxford University Press,
Oxford.
Whittaker, R. J., Triantis, K. A. And Ladle, R. J. (2008) A general dynamic
theory of oceanic island biogeography. Journal of Biogeography, 35, 977-994.
Wilson, R. R. and Kaufmann, R. S. (1987) Seamount biota and biogeography.
American Geophysical Union Geophysical Monographs, 43, 355-378.
11
Not to be cited without prior reference to the author
Table 1. Depth and age of fourteen seamounts in the Gulf of Alaska. Figures
in italics and names with an asterisk define those seamounts that have peaks
and all or part of their plateaus within 500 m of the sea surface. From Chaytor
et al., (2007).
Seamount
Bowie*
Base
depth m
2800
Peak
depth m
34
Summit
area km2
26
Min
plateau
depth, m
200
Max.
plateau
depth, m
250
Denson
3100
946
109
950
1250
Dickins*
Welker
Pratt
3000
3300
3700
419
759
719
18
104
229
750
730
910
900
Giacomini
3900
648
74
680
730
Ely
Cobb*
Warwick*
Scott
Campbell
Murray
Patton*
Marchand
3900
3000
3400
4000
3900
3300
3900
4000
2129
34
489
1005
1029
572
160
1665
9
61
42
19
3
24
2
2200
120
500
1080
1100
600
330
-
2250
260
650
1150
1200
700
430
1885?
Age
Ma
≤0.7
16.819.7
2-4
14.9
18
19.921
?
3.3
6.9
?
?
27.6
29.7
26
12
Not to be cited without prior reference to the author
Figures
Sea surface
500 metre depth
Figure 1. Seamounts as underwater islands. If the 500 m depth level is assumed to
be a boundary between the eupohotic and abyssal zones, then the peaks of
seamounts above the demarcation depth can be considered as islands. Solid arrows
signify potential exchange of taxa between seamount islands within a group. Dotted
arrows represent taxa coming in from other seamounts or from shelf areas of
equivalent depth.
13
Not to be cited without prior reference to the author
Figure 2. The basics of MacArthur and Wilson’s (1967) theory of island
biogeography predicting the number of species to be found on an island given
the rates of immigration and extinction. Inear and Ifar are the immigration rates
for islands that are near to and far from a landmass. Esmall and Elarge are the
extinction rates for small and large islands. Sx are the equilibrium numbers of
species for the different combinations x of immigration and extinction.
14
Not to be cited without prior reference to the author
Figure 3. A graphical representation of the general dynamic model as
proposed by Whittaker et al (2008). As in Figure 2, I and E represent the
immigration and extinction rates of species. S is the rate of speciation, K is the
carrying capacity and R is the realised species richness resulting from a
combination of immigration, speciation and extinction. Unlike a terrestrial
island, underwater seamount islands would reduce in height mainly through
subsidence with little erosion. Undersea landslips could still be possible and
there is evidence for these. (From Whittaker et al., (2008)).
15