| Rapid
speciation
Understanding the Lake Malawi cichlid radiation
by Patrik Bylund
Introduction
The cichlid family of fishes is one of the most species-rich of all
vertebrate families. Most of these species occur in three East African
lakes, Lake Victoria, Lake Tanganyika and Lake Malawi. Lake Malawi is
a rift lake with a maximum depth of 785 meters that is permanently stratified
below 250 meters (Love-McConnell 1993). Lake Malawi is inhabited by
9 different families of fishes (Love-McConnell 1993), but the dominant
family is Cichlidae. At least 500 endemic species of cichlid fishes
are recognized in Lake Malawi (Barlow 2000). The Lake Malawi basin is
4.5-8.6 million years old (Martens 1997). The cichlids are thought to
have invaded Lake Malawi from a monophyletic origin (Meyer 1993). These
features have made the cichlids one of the favourite model systems for
adaptive radiation. Adaptive radiation can be defined as the evolution
of ecological and phenotypic diversity within a rapidly multiplying
lineage (Schluter 2000). In this paper I will try to discuss the main
processes that have lead to the extraordinary species richness shown
by the Lake Malawi cichlids.
|
Colour differences between closely related
Aulonocara..........photo: Fredrik Hagblom
|
The
three stage radiation
Streelman and Danley (2003) put forth a general model of vertebrate
evolution, consisting of three major stages. The first stage consists
of divergence of lineages into different major habitats. In the
second radiation morphological specialization occurs, leading
to trophic differentiation within habitats. The final stage consists
of sensory communication diversification.
This model emphasises on the importance
of intrinsic factors. Streelman and Danley (2003) claims that
“the stages of most radiations occur because of some form
of selection, natural and/or sexual”. Extrinsic factors
are not unimportant, but rather shape the radiation, for example
the complete desiccation of Lake Victoria 12500 years ago, and
the recent partial desiccation of southern parts of Lake Malawi
(Kornfield and Smith 2000). The most important extrinsic factors
are minor or major lake-level fluctuations, minor fluctuations
create dynamic conditions in seemingly stable littoral communities,
while major fluctuations may cause a lake to be separated into
isolated basins (Martens 1997).
Major habitat
diversification
The first major branching of the cichlids in Lake Malawi occurred
some 700 000 years ago, based on mtDNA sequence divergence (Meyer
1993). Kornfield and Smith (2000) suggests that the age of the
branching “may be on the order of 1 my”. |
Two major clades appeared: the rock-dwelling
mbunas and another group consisting of species inhabiting sandy environments.
These two clades numbers about 200 species each. In addition to these
clades pelagic genera, such as Rhampochromis and Diplotaxodon, also
occurred. Knowledge of the pelagic and sand-dwelling cichlids in Lake
Malawi is very sparse. The major focus has been on the mbuna clade.
Mbunas are generally similar in body shape and habitat preference. However,
great differences exist in trophic morphology and feeding preferences
of the mbuna cichlids.
This pattern of habitat diversification is
consistent with other systems, for example the Artic charr Salvelinus
alpinus, in Icelandic, Norwegian and Swedish lakes. In these lakes,
where Arctic charr are occurring allopatric, between two and four different
morphs are discernible (Jonsson and Jonsson 2001). These morphs vary
in colouration, morphology, behaviour, life history and genetic characteristics.
One or two morphs are epibenthic zoobenthos feeders, one is a limnetic
planctivore and one is piscivorous (Jonsson and Jonsson 2001).
Pseudotropheus sp."elongatus mpanga" of Lake Malawi.......Photo:
Fredrik Hagblom
Trophic diversification
The second radiation stage consists of trophic diversification. This
is well studied in the Mbuna clade of Lake Malawi cichlids, for example
Albertson et al (2003) and Bootsma et al (1996). Albertson et al suggests
that the rapid trophic radiation is the result of directional selection
on the oral jaw. This rapid evolution of the feeding apparatus has been
made possible by two main adaptive innovations occurring in all cichlid
lineages, originally proposed by Liem in 1973 (Galis and Metz 1998).
First, the food processing role switched from the oral jaw to the pharyngeal
jaw. Freed from the double task of both collecting and processing the
food, the oral jaw could be specialized to different food collection
tasks. Liem’s key-innovation hypothesis can be used to explain
the proliferation of mbuna lineages during this second stage of radiation
(Galis and Metz 1998).
The directional selection experienced explains how resource competition
can drive species or populations to exploit new resource types (Schluter
2000).
Bootsma et al (1996), in contrasts with earlier stomach analyses, presents
evidence that food partitioning indeed exists among mbunas. The difference
between Bootsma et al’s stable isotope analyses and earlier stomach
contents examinations can be explained by the fact that stomach analyses
only provides a snap-shot of the feeding habits, whereas stable isotope
analyses represents a spatial and temporal integration of the feeding
habits.
The different results might indicate that mbunas which seemingly have
the same dietary preferences are spatially partitioned (Bootsma 1998).
This narrow partitioning of food resources is expected to generate a
rapid diversification of new species (Danley and Kocher 2001).
This second radiation has resulted in the
10-12 genera of the mbuna cichlids presently recognized. These genera
definitions are based on trophic morphology differences such as tooth
structures and jaw shapes (Danley and Kocher 2001).
Cichlid fishes are again not the only systems where trophic diversification
has occurred. Once more, Arctic charr can be used as an example. The
limnetic charr morph has in many lakes split into one piscivourous and
one planktivorous morph. In Thingvallavatn, Iceland has the epibenthic
morph diverged into two different morphs (Jonsson and Jonsson 2001).
Sexual selection
The third radiation, explained by sexual selection is the most controversial
in this model. Seehausen and van Alphen (1998) demonstrated that in
Lake Victoria cichlids, species-assortative mate choice existed among
females when colour differences were visible. However when the lightning
conditions were changed to monochromatic light, making it impossible
for females to distinguish colour differences, they showed non-assortative
mate choice. Instead females of both species chose the larger, more
displaying male (Seehausen and van Alphen 1998).
Seehausen et al (1997) found a correlation
between the transparency of the water and whether or not the cichlids
occurring in the basin had formed species flocks. The species flocks
had occurred in the more transparent basins. They also found that within
Lake Victoria, which exhibits a variation in water turbidity, more colourful
(red and blue) cichlids were more abundant where light and water conditions
enhanced the effect of their colour signals. The main predators of the
cichlids; cormorants, egrets and otters, spot the cichlids more easily
in transparent waters than turbid. Cormorants also predominantly catch
brightly coloured cichlids (Seehausen et al 1997). Thus the observation
that cichlids are brighter and more colourful in transparent waters
is opposed to what would be predicted by a natural selection hypothesis.
Seehausen et al (1997) suggest that the more plausible explanation to
the observation is that colouration is determined by sexual selection.
More recently Seehausen et al (2003) supported the three step radiation
model by Streelman and Danley (2003) and Danley and Kocher (2001). Seehausen
et al (2003) suggest that the great colour variation among the rapidly
radiated mbunas can be partitioned into a relative small number of core
patterns that are similar between genera. Over 2000 polymorphic loci
in 59 individuals were measured for genetic similarity in individuals
within and between populations, species and genera (Seehausen et al
2003). This resulted in the first larger species level phylogeny for
mbunas. Seehausen et al (2003) came to the conclusion that the diversity
within genera had arisen throug replicated radiations into the same
colour types, resulting in phenotypically different, but closely related
species within a region. Between regions, species with very similar
colouration, but unrelated to each other were found. This supports divergent
sexual selection during speciation, but is not consistent with models
focusing on colonization and character displacement. Kornfield and Smith’s
(2000) statement that the tertiary radiation in Lake Malawi cichlids
is extremely recent fits well with Seehausen et al’s (2003) result.
Pseudotropheus demasoni of Lake Malwai ...........Photo:Fredrik Hagblom
Conclusions and complications
The earlier models focusing on extrinsic factors, such as isolation
leading to allopatry, are not in favour any longer (Galis and Metz 1998).
Liems key-innovation hypothesis cannot be the sole factor involved explaining
the rapid divergence of Lake Malawi cichlids (Danley and Kocher 2001).
The cichlids share this key-innovation with all other fishes of the
suborder Labroidei, including wrasses and damselfishes (Barlow 2000).
If Liems hypothesis could be the sole explanation, then one would expect
the neo-tropical cichlids to exhibit the same radiation.
Many hypothesises have focused on the possibility
that sexual selection, in the form of Fisherian run-away sexual selection
or good-genes models, could explain the rapid divergence (Danley and
Kocher 2001). Galis and Metz (1998) points out that if only sexual selection
acting on male nuptial colour differences produced the radiation, ecologically
indistinguishable species would be the result and that would lead to
extinction of species in a random process much like genetic drift.
What is needed is an integrative model, that combines these models into
one that can address the entire process, and both Danley and Kocher
(2001) and Streelman and Danley (2003) present the three stage radiation
to do just that.
While this three stage model fits nicely with the data from Lake Malawi
cichlids, it appears that most other examples of adaptive radiations
among vertebrates don’t show these three stages as clearly. The
Lake Victoria cichlid fishes lack the second step of morphologic radiation
(Streelman and Danley 2003). This could be interpreted such that in
the cichlid radiation in Lake Victoria, the stages have been inverted,
with sexual selection being a stronger force than differences in trophic
morphology. However, while Lake Victoria cichlids lack the morphological
jaw specialization, they still present a striking diversity of feeding
niches, and sibling species are always characterized by small differences
in feeding behaviour (Galis and Metz 1998).
Barlow (2000) makes a vital point why the cichlids of Malawi have succeeded
to radiate to fill niches usually comprised of several families of fishes:
because of their marine origin they reproduce year-round. Most other
fishes have to reproduce in streaming water, or the lack of oxygen will
kill the eggs (Barlow 2000). Cichlids don’t have this problem
because of their parental care, whether they are mouth-brooders or substrate-brooders,
they provide the eggs with moving and thus oxygenated water. Hence cichlids
are well suited to lacustrine environments, and might be competitive
superior to other fish families in these environments.
So the question is no longer why there are
so many cichlid species, but rather, why aren’t there more other
species? Streelman and Danley (2003) propose that different constraints
on the diversification process might exist. The three main constraints
would be environmental, evolutionary history and genetic constraints.
Environmental constraints can be that the lack of size on and number
of habitats available limits the radiation. Example can once again be
taken from the Arctic charr.
Evolutionary history constrains can be described as that the existing
phenotype sets the framework in which the evolutionary processes can
work.
Genetic constraints can occur because certain genetic patterns, that
maximized divergence at one stage, limit divergence on another stage
(Streelman and Danley 2003)
References
Albertson, R. Craig, Streelman, J. Todd and
Kocher, Thomas D (2003) Directional selection has shaped the oral jaws
of Lake Malawi cichlid fishes. Proceedings of the National Academy of
Sciences of the USA 100 5252-5257
Bootsma, Harvey A., Hecky, Robert E., Hesslein,
Ray H. and Turner, George F. (1996) Food partitioning among Lake Malawi
nearshore fishes as revealed by stable isotope analysis. Ecology 77
1286-1290
Barlow, George W (2000), The Cichlid Fishes:
Natures Grand Experiment In Evolution. Perseus Publishing
Danley, Patrick D. and Kocher, Thomas D. (2001),
Speciation in Rapidly Diverging Systems: Lessons from Lake Malawi. Molecular
Ecology 10 1075-1086
Galis, Frietson and Metz, Johan A. J. (2000)
Why are there so many cichlid species? Trends in Ecology and Evolution
13 1-2
Jonsson, B. and Jonsson N. (2001) Polymorphism
and speciation in Arctic charr. Journal of Fish Biology 58 605-638
Kornfield, Irv and Smith, Peter F. (2000)
African Cichlid Fishes: Model Systems for Evolutionary Biology. Annual
review of ecology, evolution and systematics 31 163-196
Love-McConnell, Rosemary (1993), Fish Faunas
of the African Great Lakes: Origins, Diversity and Vulnerability. Conservation
Biology 7 634-643
Martens, Koen (1997), Speciation in Ancient
Lakes. Trends in Ecology and Evolution 12 177-182
Meyer, Axel (1993) Phylogenetic Relationships
and Evolutionary Processes in East African Cichlid Fishes. Trends in
Ecology and Evolution 8 279-284
Schluter, Ralph (2000) The ecology of Adaptive
Radiation Oxford University Press
Seehausen, Ole, van Alphen, Jacques J. M.
and Witte, Frans (1997) Cichlid Fish Diversity Threatened by Eutrophication
That Curbs Sexual Selection. Science 277 1808-1811
Seehausen, Ole and van Alphen, Jacques J.
M. (1998) The effect of male colouration on female mate choice in closely
related Lake Victoria cichlids (Haplochromis nyererei complex). Behavioral
ecology and sociobiology 42 1-8
Seehausen, Ole, Allender Charlotte J., Knight,
Mairi E., Turner, George F. and Maclean, Norman (2003) Divergent selection
during speciation of Lake Malawi cichlid fishes inferred from parallel
radiations in nuptial colouration. Proceedings of the National Academy
of Sciences of the USA 100 14074-14079
Streelman, J. Todd and Danley, Patrick D.
(2003), The Stages of Vertebrate Evolutionary Radiation, Trends in Ecology
and Evolution 18 126-131
|
